scieee Open visual document viewer

Posttranscriptional regulation of glutamine synthetase in the filamentous cyanobacterium Anabaena sp. PCC 7120: Differential expression between vegetative cells and heterocysts

Galmozzi, Carla V.; Saelices Gómez, Lorena; Florencio Bellido, Francisco Javier; Muro Pastor, María Isabel

Abstract

Genes homologous to those implicated in glutamine synthetase (GS) regulation by protein-protein interaction in the cyanobacterium Synechocystis sp. strain PCC 6803 are conserved in several cyanobacterial sequenced genomes. We investigated this GS regulatory mechanism in Anabaena sp. strain PCC 7120. In this strain the system operates with only one GS inactivation factor (inactivation factor 7A [IF7A]), encoded by open reading frame (ORF) asl2329 (gifA). Following addition of ammonium, expression of gifA is derepressed, leading to the synthesis of IF7A, and consequently, GS is inactivated. Upon ammonium removal, the GS activity returns to the initial level and IF7A becomes undetectable. The global nitrogen control protein NtcA binds to the gifA promoter. Constitutive high expression levels of gifA were found in an Anabaena ntcA mutant (CSE2), indicating a repressor role for NtcA. In vitro studies demonstrate that Anabaena GS is not inactivated by Synechocystis IFs (IF7 and IF17), indicating the specificity of the system. We constructed an Anabaena strain expressing a second inactivating factor, containing the amino-terminal part of IF17 from Synechocystis fused to IF7A. GS inactivation in this strain is more effective than that in the wild type (WT) and resembles that observed in Synechocystis. Finally we found differential expression of the IF system between heterocysts and vegetative cells of Anabaena.

Full text

JOURNAL OF BACTERIOLOGY, Sep . 2010, p. 4701–4711 Vol. 192, No. 18 0021-9193/10/$12.00 doi:10.1128/JB.00222-10 Copy igh © 2010, Ame ican Socie y o Mic obiology. All Righ s Rese ed. Pos ansc ip ional Regula ion o Glu amine Syn he ase in he Filamen ous Cyanobac e ium Anabaena sp. PCC 7120: Di e en ial Exp ession be ween Vege a i e Cells and He e ocys s 䌤 Ca la V. Galmozzi, Lo ena Saelices, F ancisco J. Flo encio, and M. Isabel Mu o-Pas o * Ins i u o de Bioquímica Vege al y Fo osín esis, CSIC-Uni e sidad de Se illa, Ame´ ico Vespucio 49, E-41092 Se ille, Spain Recei ed 2 Ma ch 2010/Accep ed 8 July 2010 Genes homologous o hose implica ed in glu amine syn he ase (GS) egula ion by p o ein-p o ein in e ac- ion in he cyanobac e ium Synechocys is sp. s ain PCC 6803 a e conse ed in se e al cyanobac e ial sequenced genomes. We in es iga ed his GS egula o y mechanism in Anabaena sp. s ain PCC 7120. In his s ain he sys em ope a es wi h only one GS inac i a ion ac o (inac i a ion ac o 7A [IF7A]), encoded by open eading ame (ORF) asl2329 (gi A). Following addi ion o ammonium, exp ession o gi A is de ep essed, leading o he syn hesis o IF7A, and consequen ly, GS is inac i a ed. Upon ammonium emo al, he GS ac i i y e u ns o he ini ial le el and IF7A becomes unde ec able. The global ni ogen con ol p o ein N cA binds o he gi A p omo e . Cons i u i e high exp ession le els o gi A we e ound in an Anabaena n cA mu an (CSE2), indi- ca ing a ep esso ole o N cA. In i o s udies demons a e ha Anabaena GS is no inac i a ed by Synecho- cys is IFs (IF7 and IF17), indica ing he speci ici y o he sys em. We cons uc ed an Anabaena s ain exp essing a second inac i a ing ac o , con aining he amino- e minal pa o IF17 om Synechocys is used o IF7A. GS inac i a ion in his s ain is mo e e ec i e han ha in he wild ype (WT) and esembles ha obse ed in Synechocys is. Finally we ound di e en ial exp ession o he IF sys em be ween he e ocys s and ege a i e cells o Anabaena. Glu amine syn he ase (GS) ca alyzes he ATP-dependen ami- da ion o glu ama e o yield glu amine. This enzyme ope a es sequen ially wi h he enzyme glu ama e syn hase (GOGAT), which ca alyzes he ans e o he amide g oup om glu amine o 2-oxoglu a a e o yield wo molecules o glu ama e. This pa hway (commonly known as he GS-GOGAT cycle) ep e- sen s he connec ing s ep be ween ca bon and ni ogen me ab- olism. In mos o he sys ems s udied, con ol o GS ac i i y esponds o ca bon and ni ogen signals. In he p esence o abundan ca bon sou ces, ni ogen de iciency esul s in a high le el o GS ac i i y. In con as , when he ni ogen sou ce is abundan , GS ac i i y is down egula ed (12, 14). In cyanobac e ia, GS ype I (he e e e ed o as GS) is modula ed a he ansc ip ional and pos ansc ip ional le els, depending on he ca bon and ni ogen supply (19). The pos - ansla ional modi ica ion by adenylyla ion ha occu s in en- e obac e ial glu amine syn he ase does no exis in cyanobac- e ia. Howe e , in hese o ganisms an ammonium-dependen GS pos ansla ional egula ion mechanism in ol ing p o ein- p o ein in e ac ion has been epo ed (8, 28). This sys em has been s udied in de ail in he cyanobac e ium Synechocys is sp. s ain PCC 6803 and consis s o a e e sible in e ac ion o GS wi h wo small p o eins, inac i a ion ac o 7 (IF7) and IF17, encoded by he gi A and gi B genes, espec i ely (8). The anal- ysis o mu an s ains de oid o IF7, IF17, o bo h e ealed ha each o hese p o eins con ibu es o GS inac i a ion in i o, and a maximal le el o inac i a ion was obse ed when bo h p o eins we e p esen (8). The exp ession o gi A and gi B genes, encoding IF7 and IF17, espec i ely, is ep essed by N cA, he main ac o esponsible o ni ogen con ol in cya- nobac e ia (9, 10). Thus, when ammonium is added o he medium, IF7 and IF17 a e exp essed and GS is inac i a ed. Ammonium emo al p o okes ep ession o gi genes and also de e mines he apid deg ada ion o IF7 and IF17 p e iously accumula ed upon ammonium addi ion (6). In ilamen ous cyanobac e ia, ea ly s udies by O and Ha- selko n demons a ed ha GS om Anabaena sp. s ain PCC 7120 is con olled nei he by adenylyla ion no by eedback inhibi ion by glu amine; howe e , le els o glu amine syn- he ase a e lowe in ammonium-g own cells han in cells g own using ni a e o dini ogen as he ni ogen sou ce (25, 26). As in o he cyanobac e ia, exp ession o he s uc u al gene o glu amine syn he ase (glnA) is egula ed a he ansc ip ional le el in Anabaena and his con ol is media ed by N cA (5). The p omo e egion o he glnA gene has a complex s uc u e in his s ain and has been well cha ac e ized unde di e en ni ogen egimens in ege a i e cells and he e ocys s (31). Genes homologous o gi A o gi B ha e been ound in many cyanobac e ial genomes, al hough hey seem o be absen in P ochlo ococcus. Howe e , ammonium-p omo ed down egula- ion o glu amine syn he ase ac i i y has been well documen ed only in Synechocys is sp. PCC 6803. The e o e, we ound i in e es ing o explo e he ope a ion o he GS egula o y mech- anism media ed by inac i a ing ac o s (IF7 o IF17 homologs) in o he cyanobac e ial g oups. Anabaena sp. PCC 7120 possesses a single e edoxin-dependen GOGAT (Fd-GOGAT) enzyme, whe eas Synechocys is ha bo s bo h NADH-dependen and e e- doxin-dependen GOGAT enzymes (16, 22). In addi ion, Fd- GOGAT is absen in he e ocys s om Anabaena, indica ing * Co esponding au ho . Mailing add ess: Ins i u o de Bioquímica Veg- e al y Fo osín esis, Ame´ ico Vespucio 49, E-41092 Se ille, Spain. Phone: 34-954-489573. Fax: 34-954-460065. E-mail: [email p o ec ed]. 䌤 Published ahead o p in on 16 July 2010. 4701 on July 20, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://jb.asm.o g/Downloaded om he lack o a comple e GS-GOGAT pa hway in hese cells (16). Hence, we decided o u he in es iga e GS egula ion and ammonium sensing in bo h cell ypes om his model cya- nobac e ium. He e we demons a e ha GS om a ilamen- ous cyanobac e ium is also egula ed pos ansc ip ionally by he IF-media ed sys em, which is N cA dependen . We also analyze he speci ici y o he in e ac ion be ween IFs and GSs om di e en s ains. Fu he mo e, using g p as a epo e gene, we show a di e en ial ammonium sensing be ween he - e ocys s and ege a i e cells. MATERIALS AND METHODS S ains and cul u e condi ions. Anabaena sp. PCC 7120 and Synechocys is sp. PCC 6803 wild- ype (WT) s ains and he Anabaena s ains gene a ed in his wo k, he ⌬gi A, ACHI, and AGFP s ains, we e g own pho oau o ophically a 30°C in BG11 medium (29) supplemen ed wi h 1 g li e ⫺1 NaHCO 3 (BG11C) and bubbled wi h a con inuous s eam o 1% ( ol/ ol) CO 2 in ai unde con in- uous luo escen illumina ion (50 ␮mol pho ons 䡠m ⫺2 䡠s ⫺1 whi e ligh ). The CSE2 s ain was cul i a ed in BG11C medium supplemen ed wi h 5 mM NH 4 Cl and 10 mM N- is(hyd oxyme hyl)-me hyl-2-aminoe hanesul onic acid (TES) bu e , pH 7.5. The mosynechococcus elonga us BP-1 was g own unde he same condi ions desc ibed abo e o Anabaena and Synechocys is bu a 45°C. Fo pla e cul u es, BG11C liquid medium was supplemen ed wi h 1% (w / ol) aga . Am- monium ea men o cul u es was pe o med by addi ion o 10 mM NH 4 Cl and 20 mM TES bu e , pH 7.5. Ammonium emo al was ca ied ou by ha es ing he cells by il a ion, washing hem, and esuspending hem wi h BG11C. To place cul u es unde dini ogen g ow h condi ions, cells om BG11C medium o om BG11C supplemen ed wi h NH 4 Cl we e ha es ed by il a ion a oom empe a u e, washed, and esuspended in BG11 0 C medium (BG11C medium wi hou NaNO 3 ). Inse ional mu agenesis o he gi A gene in Anabaena. Two agmen s o 575 and 400 bp, encompassing pa o he asl2329 locus and he 5⬘ egion and pa o he asl2329 locus and he 3⬘ egion, espec i ely, we e ampli ied by PCR using Anabaena genomic DNA. These agmen s we e cloned in o pGEM-T plasmid (P omega), gene a ing he plasmid pAN3. This plasmid con ains a dele ion o he asl2329 locus and also a BamHI es ic ion si e. This si e was used o clone aSm Sp C.S3 casse e (27) om pRL463 (pUC18/19 con aining L.HEH1 and C.S3; nomencla u e o Elhai and Wolk [4]) in bo h o ien a ions, gene a ing plasmids pANSP(⫹) and pANSP(⫺), espec i ely. XhoI-diges ed agmen s om pANSP(⫹) o pANSP(⫺) we e liga ed o XhoI-diges ed pRL278 ec o (1), gene a ing he a ge ing plasmids pRLANSP(⫹) and pRLANSP(⫺), espec- i ely. To gene a e ⌬gi A(⫹) and ⌬gi A(⫺) s ains, plasmids pRLANSP(⫹) and pRLANSP(⫺), espec i ely, we e in oduced in o he Anabaena wild- ype s ain by conjuga ion (3). Subs i u ion o wild- ype gi A by C.S3-in e up ed e sions was con i med by Sou he n blo analysis. Sequencing o gi A locus. The gi A locus was PCR ampli ied om genomic DNA o wild- ype Anabaena and he CSE2 mu an s ain by using p ime s 5⬘CTCTTGCAGTGTTCTGTTGCTGG3⬘and 5⬘GAGTTACTTCCTCTAATA ACAACC3⬘. Di ec sequencing o PCR p oduc s was ca ied ou by he Eu o ins MWG ope on sequencing se ice. Conjuga ion o wild- ype and mu a ed gi A o he CSE2 s ain. The gi A wild- ype e sion was ampli ied by PCR using p ime s 5⬘GATCAGATCTCTC TTGCAGTGTTCTGTTGCTGG3⬘and 5⬘GATCAGATCTGGAAGTAACTT CAACAATGAG3⬘and Anabaena DNA as empla e. The gi A mu a ed e sion was gene a ed by a wo-s ep PCR p ocess using p ime s 5⬘GATGCGCTAATA TTAGCAAGTGAAGAATCG3⬘and 5⬘CGATTCTTCACTTGCTAATATTA GCGCATC3⬘ o in oduce mu a ions, and he same p ime s we e used o am- pli y he wild- ype e sion in he second s ep. F agmen s con aining bo h gi A e sions we e BglII diges ed and liga ed o BglII-diges ed pCSAV81 (31), gen- e a ing plasmids pWTgi A and pMTgi A. These plasmids we e ans e ed by conjuga ion (3) o s ain CSE2. The co ec in eg a ion o hese cons uc s in he nucA-nuiA egion o he Anabaena ␣megaplasmid was checked by PCR. GS assay. GS ac i i y was de e mined in si u by using he Mn 2⫹ -dependen ␥-glu amyl ans e ase assay in cells pe meabilized wi h mixed alkyl ime hylam- monium b omide (MTA) (17). Fo he analysis o he in i o GS-IF in e ac ion, binding eac ions we e ca ied ou in a inal olume o 20 ␮l con aining pu i ied Anabaena o Synechocys is GS and inc easing amoun s o IF7, IF17, IF7A, o IF17N/IF7A in HEPES-NaOH bu e , pH 7.0, 50 mM KCl. A e he GS-IF complex o ma ion, he same GS assay desc ibed abo e bu wi hou MTA addi- ion was pe o med. One uni o GS ac i i y co esponds o he amoun o enzyme ha ca alyzes he syn hesis o 1 ␮mol min ⫺1 o ␥-glu amylhyd oxama e. RNA isola ion and No he n blo analysis. To al RNA was isola ed om 25-ml samples o Anabaena cul u es a he mid-exponen ial phase (3 o 5 ␮g/ml chlo- ophyll). Ex ac ions we e pe o med by o exing cells in he p esence o phe- nol-chlo o o m and acid-washed baked glass beads (0.25- o 0.3-mm diame e ; B aun, Melsungen, Ge many) as p e iously desc ibed (7). Fo No he n blo ing, 15 ␮g o o al RNA was loaded pe lane and elec opho esed on dena u ing o maldehyde-con aining 1.2% aga ose gels. T ans e o nylon memb anes (Hy- bond N-plus; Ame sham Pha macia Bio ech), p ehyb idiza ion, hyb idiza ion, and washes we e pe o med as ecommended by he manu ac u e . PCR-syn- hesized agmen s encompassing he en i e gi A,gi B,o glnA genes we e used as p obes. As a con ol, he il e s we e ep obed wi h a 640-bp DNA agmen con aining he cons i u i ely exp essed RNase P RNA gene ( npB) om Anabaena (33). Hyb idiza ion signals we e quan i ied wi h a Cyclone Phospho sys em (Packa d). P o ein exp ession and pu i ica ion. Fo IF7A-His 6 , IF7-His 6 , IF17-His 6 , and IF17N/IF7A-His 6 exp ession, NdeI-XhoI agmen s con aining he co espond- ing gi gene we e syn hesized by PCR and cloned in o he pET24a(⫹) plasmid (No agen, La Jolla, CA) o gene a e pASET24, pSET24, pLET24, and pALSET24 plasmids, espec i ely. Cons uc ion o he gi B/gi A chime ic gene o IF17N/IF7A p o ein exp ession is desc ibed below. Exponen ially g owing Esch- e ichia coli BL21 cells ans o med wi h each o hese plasmids we e ea ed wi h 0.5 mM isop opyl-␤-D- hiogalac oside o 4 h. IF17-His 6 was pu i ied om E. coli as p e iously desc ibed (6). IF7A-His 6 , IF7-His 6 , and IF17N/IF7A-His 6 we e pu i ied by Ni-a ini y ch oma og aphy using His-Bind ma ix (No agen) ollow- ing he manu ac u e ’s ins uc ions. F ac ions ha showed GS inac i a ion ac- i i y we e pooled and subjec ed o gel il a ion ch oma og aphy using a HiLoad 16/60 Supe dex 75 gel il a ion column (GE Heal hca e) unning on an Ak a as p o ein liquid ch oma og aphy (FPLC) sys em. Fo Anabaena GS exp ession, a NdeI-BamHI agmen including a his idine- agged modi ied e sion o he glnA gene was syn hesized by PCR and cloned in o he pET-3a plasmid (No agen, La Jolla, CA) o gene a e pAGS. Exponen- ially g owing E. coli BL21 cells ans o med wi h he pAHGS plasmid we e ea ed wi h 0.5 mM isop opyl-␤-D- hiogalac oside o 4 h. Anabaena His-GS was pu i ied by Ni-a ini y ch oma og aphy using His-Bind ma ix (No agen) ollow- ing he manu ac u e ’s ins uc ions. F ac ions ha showed GS ac i i y we e pooled and subjec ed o gel il a ion using a HiLoad 16/60 Supe dex 200 gel il a ion column (GE Heal hca e) unning on an Ak a FPLC sys em. Fo Synechocys is GS exp ession, a p e iously desc ibed his idine- agged mod- i ied e sion o he glnA gene (8) was cloned in o pBluesc ip SK(⫹) in he same o ien a ion as he plac p omo e o gene a e he pSHGS plasmid. Synechocys is His-GS was pu i ied om E. coli DH5␣cells ans o med wi h pSHGS using he same me hods desc ibed abo e o Anabaena His-GS. Fo Anabaena N cA exp ession, a NdeI-XhoI agmen encompassing he en i e n cA gene was syn hesized by PCR and cloned in o he pET24a(⫹) plasmid (No agen, La Jolla, CA) o gene a e pAN cA. Exponen ially g owing E. coli BL21 cells ans o med wi h pAN cA plasmid we e ea ed wi h 0.5 mM isop opyl-␤-D- hiogalac oside o 4 h. Anabaena N cA was pu i ied by Ni-a ini y ch oma og aphy using His-Bind ma ix (No agen) ollowing he manu ac u e ’s ins uc ions. Fo u he pu i ica ion, he sample was subjec ed o gel il a ion ch oma og aphy using a HiLoad 16/60 Supe dex 75 gel il a ion column (GE Heal hca e) unning on an Ak a FPLC sys em. An i-IF7A an ibody p oduc ion and Wes e n blo ing. An i-IF7A an ise um was ob ained acco ding o s anda d immuniza ion p o ocols by injec ing pu i ied IF7A-His 6 in o abbi s. Pu i ied polyclonal an ibodies ob ained agains Synecho- coccus sp. s ain PCC 6301 glu amine syn he ase (15) we e used o de ec Anabaena glu amine syn he ase. An ibodies ob ained agains Synechocys is sp. PCC 6803 hio edoxin A (T xA) we e used o de ec Anabaena T xA. Fo Wes e n blo analysis, p o eins we e ac iona ed by 15% SDS-PAGE acco ding o he me hod o Laemmli (11) and immunoblo ed wi h an i-IF7A (1:2,000), an i-T xA (1:3,000), o an i-GS (1:15,000). The ECL Plus immunoblo ing sys- em (Ame sham) was used o de ec he di e en an igens wi h an i- abbi seconda y an ibodies conjuga ed o ho se adish pe oxidase (1:12,000). Enhanced chemiluminescence (ECL) signals we e quan i ied using a ChemiDocXRS ap- pa a us (Bio-Rad, He cules, CA) and he Quan i yOne p og am. P ime ex ension analysis. Oligonucleo ide PEIF1 (5⬘GATATTGGCGCAT CATAATGG3⬘; om nucleo ide 47 o 23 o he coding egion), end labeled wi h T4 polynucleo ide kinase and [␥- 32 P]dATP (3,000 Ci mmol ⫺1 ), was used o p ime ex ension analysis o gi A. Annealing and ex ension eac ions using o al RNA om Anabaena we e pe o med as p e iously desc ibed (9). Ex ension p oduc s we e analyzed on a polyac ylamide sequencing gel oge he wi h a sequencing eac ion mix u e o he gi A 5⬘ egion using PEIF1 oligonucleo ide. 4702 GALMOZZI ET AL. J. BACTERIOL. on July 20, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://jb.asm.o g/Downloaded om Gel e a da ion assays. N cA-His 6 , exp essed and pu i ied as desc ibed abo e, was used in gel e a da ion assays. DNA agmen s we e end labeled wi h [␣- 32 P]dCTP using Sequenase e sion 2.0 enzyme. The binding eac ions and elec opho esis we e ca ied ou as p e iously desc ibed (21). The Pgi A1 p omo e p obe was ob ained by BglII diges ion o a PCR-ampli ied agmen using oligonucleo ides N cA1 (5⬘CTAGCGGCCGCAGTGTTCTGTTGC3⬘) and N cA2 (5⬘GCGCAGATCTCCTATG3⬘). The Pgi A2 p omo e p obe was ob ained by No I diges ion o a PCR-ampli ied agmen using oligonucleo ides N cA2 and N cA3 (5⬘CTAGCGGCCGCAATTACATAAGTATTACA3⬘). Gene a ion o he ACHI Anabaena s ain. To gene a e a chime ic gene be- ween gi B om Synechocys is and gi A om Anabaena, unde he con ol o he gi B p omo e , wo o e lapping DNA agmen s we e ampli ied by PCR. A agmen con aining he gi B p omo e and pa o he coding egion was am- pli ied om Synechocys is genomic DNA using oligonucleo ides P17EcoRI (5⬘C ATCCAGCCCGAATTCCATCTCCCTCG3⬘) and A17NH (5⬘ATAGACATTT GGCTGGGAGCCGCAGCGAC3⬘). Ano he agmen , con aining he gi A coding egion and pa o he 3⬘ egion, was ampli ied om Anabaena genomic DNA using oligonucleo ides AIFNH (5⬘CCAGCCAAATGTCTATTCAAGAA AAATCTCG3⬘) and AIFPs I (5⬘GATCCTGCAGGGAAGTAACTTCAACAA TGAG3⬘). The chime ic gene was PCR syn hesized om hese wo agmen s and cloned a e EcoRI/Ps I diges ion in o pCSEL24 plasmid (23), diges ed wi h he same enzymes, ende ing pACHI. This plasmid was in oduced in o Anabaena by conjuga ion (3). The co ec in eg a ion o his cons uc in he nucA-nuiA egion o he Anabaena ␣megaplasmid was checked by PCR. Gene a ion o AGFP Anabaena s ain and isualiza ion o g een luo escen p o ein (GFP). A p omo e less g p gene was PCR ampli ied om pCSEL19 (18) wi h p ime s 5⬘CTAGGACTGTATGTCTAAAGGAGAAGAAC3⬘and 5⬘CTA GGACTGTACGTCTTATTTGTATAGTTCATCCATGC3⬘. This agmen was AhdI diges ed and liga ed o pANSP(⫺) plasmid (desc ibed abo e), con aining he Anabaena gi A genomic egion wi h a dele ion o he asl2329 open eading ame (ORF), diges ed wi h he same enzyme. The esul ing plasmid, pAGFP1, con ains a ansla ional usion be ween he gi A p omo e egion and p omo e - less g p. A XhoI-diges ed agmen om pAGFP1 was liga ed o XhoI-diges ed pRL278 ec o (1), gene a ing he a ge ing plasmid pAGFP2. To gene a e he AGFP s ain, pAGFP2 was in oduced in o he Anabaena wild- ype s ain by conjuga ion (3). These plasmids can be in eg a ed upon homologous ecombi- na ion in he gi A locus o Anabaena. The co ec in eg a ion was con i med by Sou he n blo analysis. The accumula ion o he GFP epo e was analyzed by lase con ocal mic os- copy as desc ibed p e iously (18). RESULTS Compa a i e GS inac i a ion/ eac i a ion p ocesses in di - e en cyanobac e ia. Taking in o accoun ha ORFs homol- ogous o he Synechocys is gi A and gi B genes a e p esen in se e al cyanobac e ium sequenced genomes, we decided o examine i he addi ion o ammonium causes inac i a ion o GS in o he membe s o he phylum, as desc ibed o Synecho- cys is (8). Figu e 1 shows compa a i e kine ics o his p ocess in he he e ocys - o ming, model cyanobac e ium Anabaena sp. PCC 7120, he he mophilic s ain The mosynechococcus elon- ga us BP-1, and Synechocys is sp. PCC 6803. Addi ion o am- monium p o okes a quick d op in GS ac i i y in he The mo- synechococcus and Synechocys is s ains; howe e , in Anabaena he p ocess is slowe , and by 6 h ollowing ammonium addi ion, GS ac i i y eaches only abou 60% o he ini ial le el. Am- monium emo al leads o a apid eco e y o GS ac i i y in all he s ains analyzed (Fig. 1). Dele ion mu an s o asl2329 ORF om Anabaena sp. PCC 7120 a e impai ed in GS inac i a ion. The Anabaena asl2329 ORF sha es homology wi h bo h GS inac i a ion ac o s om Synechocys is (IF7 and IF17) (8). To es whe he his ORF is in ol ed in Anabaena GS inac i a ion, we cons uc ed dele ion mu an s o his gene (Fig. 2A) and in es iga ed he ammonium- dependen GS inac i a ion p ocess in wild- ype (WT) Anabaena and mu an s ain cul u es. GS ac i i ies we e sim- ila in he wo s ains gene a ed [⌬gi A(⫹) and ⌬gi A(⫺) s ains]. Mo eo e , as shown in Fig. 2B, GS ac i i y shows only mino changes (less han 15%) a e ammonium addi ion in asl2329 ORF dele ion mu an s compa ed o he wild- ype s ain. These esul s demons a e ha his gene is he gi A o holog in Anabaena (8) and ha he p oduc o his ORF is a GS inac i a ion ac o (he e called IF7A). Anabaena gi A gene exp ession depends on ni ogen s a us. As a i s s ep in he cha ac e iza ion o gi A gene exp ession, we analyzed he gi A mRNA le el in pa allel wi h IF7A accu- mula ion in he ammonium-media ed GS inac i a ion/ eac i- a ion p ocesses. To moni o he cellula le el o IF7A, we p oduced speci ic an ibodies agains his p o ein. Figu e 3A shows ha ammonium addi ion o ni a e-g own cells p o okes an inc ease o gi A mRNA and ha his le el emains ele a ed du ing he ammonium ea men . Ammonium emo al esul s in a d op o gi A ansc ip , eaching he s eady-s a e le el obse ed in ni a e-g own cells wi hin 1 h. Wi h espec o he p o ein IF7A, i was unde ec able in ni a e-g own cells and accumula ed a e ammonium addi ion. Howe e , ammonium emo al led o a apid dec ease in IF7A abundance, and he p o ein was no de ec able 1 h a e ammonium elimina ion (Fig. 3B). In pa allel wi h No he n and Wes e n blo ing ex- pe imen s, we measu ed GS ac i i y o e he same ime cou se (Fig. 3C). A p ecise in e se co ela ion be ween IF abundance and GS ac i i y was obse ed. As a con ol o p o ein loading, memb anes o Wes e n blo ing expe imen s we e incuba ed also wi h an i-T xA an ibodies. Thio edoxin A (T xA) is con- s i u i ely exp essed, independen ly o he ni ogen sou ce (2). FIG. 1. Time cou se o he GS inac i a ion and eac i a ion p o- cesses in h ee di e en cyanobac e ia. Cells o Anabaena 7120, Syn- echocys is 6803, and The mosynechococcus elonga us we e g own in BG11C medium using ni a e as ni ogen sou ce. A he ime indica ed by an a ow, 10 mM NH 4 Cl was added and GS ans e ase ac i i y was de e mined in si u. An a ow also indica es he ime a which cells we e washed wi h ammonium- ee medium and GS eac i a ion ook place. One hund ed pe cen GS ac i i ies co espond o 926, 1,540, and 1,138 mU mg o p o ein ⫺1 o Anabaena,Synechocys is, and The mosynecho- coccus, espec i ely. VOL. 192, 2010 GS POSTTRANSCRIPTIONAL REGULATION IN ANABAENA PCC 7120 4703 on July 20, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://jb.asm.o g/Downloaded om Anabaena GS is no inac i a ed by he Synechocys is IFs. To cha ac e ize in i o he GS-IF7A in e ac ion, we pu i ied Anabaena GS and IF7A exp essed in E. coli. In o de o s udy he speci ici y o he GS-IF in e ac ion, we analyzed also he componen s o he p e iously cha ac e ized GS inac i a ion sys em om Synechocys is (GS, IF7, and IF17) (8). The h ee pu i ied inac i a ion ac o s (IF7, IF17, and IF7A) inhibi ed Synechocys is GS, bu only IF7A was able o inhibi Anabaena GS (Fig. 4). In addi ion, IF7A-Synechocys is GS in e ac ion seems o be s onge han IF7A-Anabaena GS in e ac ion, because equal amoun s o his inac i a ing ac o p o oke s onge inhibi ion o Synechocys is GS han o Anabaena GS (Fig. 4C). N cA egula es he Anabaena gi A p omo e . To iden i y he p omo e egion o he gi A gene (Pgi A), he ansc ip ion s a poin (TSP) o he gene was de e mined by p ime ex en- sion analysis. The gi A TSP was mapped o nucleo ide ⫺43 wi h espec o he ansla ion s a codon (Fig. 5A). Fou nucleo ides ups eam o he TSP, a pu a i e ⫺10 box in he o m TATATT was ound. No ob ious ⫺35 box was de ec ed. As obse ed in RNA blo ing expe imen s, he p ime ex en- sion p oduc om he gi A p omo e was mo e abundan in samples om cells g own in he p esence o ammonium han in samples om hose g own in he p esence o ni a e o using RNA om ni ogen- ixing cells. Taking in o accoun he in lu- ence o ni ogen s a us on he exp ession o gi A and he FIG. 2. Analysis o he Anabaena asl2329 mu an s ain. (A) Sche- ma ic ep esen a ion o he asl2329 genomic egion in he wild- ype s ain and si e o inse ion o he CS3 casse e, con aining he aadA gene in bo h o ien a ions, o gene a e ⌬gi A(⫹) and ⌬gi A(⫺) s ains, espec i ely. (B) Ammonium-dependen GS inac i a ion in Anabaena WT and ⌬gi A mu an s. Cells we e g own in BG11C medium using ni a e as ni ogen sou ce. A 10 mM concen a ion o NH 4 Cl was added (a ow), and GS ans e ase ac i i y was de e mined in si u a he indica ed imes. The cu es ep esen a i hme ic means om h ee independen expe imen s and hei s anda d de ia ion alues. FIG. 3. IF7A exp ession du ing he GS inac i a ion/ eac i a ion p ocesses. A he ime indica ed by an a ow, 10 mM NH 4 Cl was added o Anabaena cells cul i a ed wi h ni a e as ni ogen sou ce. An a ow also indica es he ime a which cells we e washed wi h ammonium- ee medium and GS eac i a ion ook place. (A) No he n blo assay o he gi A gene unde di e en ni ogen condi ions. To al RNA was isola ed om cells g own wi h ni a e (0) and a e ammonium addi- ion (⫹NH 4 ⫹ ) o emo al (⫺NH 4 ⫹ ) a he indica ed imes (min). Gels we e blo ed and hyb idized wi h he gi A p obe. The il e s we e s ipped and ehyb idized wi h an npB p obe as a loading con ol. (B) Wes e n blo assay o IFA du ing GS inac i a ion/ eac i a ion p ocesses. F om he same cul u e as ha used o No he n blo analysis, samples we e aken om ni a e-g own cells (0) and a e ammonium addi ion (⫹NH 4 ⫹ ) o emo al (⫺NH 4 ⫹ ) a he indi- ca ed imes (min). To al p o eins we e isola ed and esol ed by SDS-PAGE, blo ed, and incuba ed wi h an i-IF7A and an i-T xA an ibodies. (C) Time cou se o he GS ac i i y. Samples we e aken, du ing he inac i a ion and eac i a ion p ocesses, o de e mina- ion o GS ac i i y. 4704 GALMOZZI ET AL. J. BACTERIOL. on July 20, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://jb.asm.o g/Downloaded om p e iously desc ibed N cA-dependen ep ession o gi genes om Synechocys is sp. PCC 6803 (9), he p esence o N cA binding si es in he gi A p omo e egion was analyzed. Two consensus N cA binding si es (13) cen e ed a posi ions ⫺28.5 and ⫺77.5, espec i ely, ups eam o he gi A TSP we e ound (Fig. 5B). I is wo h no ing ha he N cA consensus si e cen e ed a posi ion ⫺28.5 wi h espec o he TSP is loca ed a exac ly he same dis ance om he ⫺10 box as is he ep essing N cA binding si e desc ibed o he gi A p omo e om Syn- echocys is (Fig. 5B). To es i N cA binds o Pgi A, elec opho e ic mobili y shi assays using pu i ied Anabaena N cA p o ein we e pe o med. Anabaena N cA was exp essed in E. coli and pu i ied as a his idine- agged e sion. Binding assays we e pe o med wi h wo DNA agmen s, Pgi A1, which spans posi ions ⫺134 o ⫹36 wi h espec o he TSP, and a sho e agmen , Pgi A2, which spans posi ions ⫺77 o ⫹36 and lacks he GTA iple o he consensus N cA binding si e cen e ed a posi ion ⫺77.5. When N cA was incuba ed wi h Pgi A1, wo N cA-DNA com- plexes we e de ec ed (Fig. 6A); howe e , when he Pgi A2 p obe was used, only one N cA-DNA complex could be de- ec ed (Fig. 6B). These esul s indica e ha N cA binds in i o o bo h consensus ecogni ion si es ound in he gi A p omo e egion. On he o he hand, 2-oxoglu a a e has been epo ed o inc ease he binding a ini y o N cA o se e al ni ogen- eg- ula ed p omo e s (32). As shown in Fig. 6A, he p esence o his me aboli e in he binding assay has a posi i e e ec on N cA ecogni ion o Pgi A. To demons a e ha he ansc ip ional egula o N cA con- ols he syn hesis o he gi A mRNA, we de e mined he le el o gi A ansc ip in he N cA mu an s ain CSE2 (5). As a con ol, we checked he le el o exp ession o glnA, he an- sc ip ion o which is posi i ely con olled by N cA (5, 31). We analyzed he s eady-s a e mRNA le els o hese genes in he wild ype and in he CSE2 s ain unde h ee di e en condi- ions: ni a e u iliza ion, ammonium u iliza ion, and ni ogen dep i a ion. Ammonium-g own CSE2 o wild- ype cells we e ans e ed o 6h oni a e- o ammonium-con aining me- dium o o ni ogen- ee medium, and samples we e aken o RNA isola ion. As p e iously epo ed, he amoun o glnA mRNA in he wild- ype s ain inc eased upon incuba ion in medium con aining ni a e o no combined ni ogen. Howe e , induc ion was se e ely impai ed in CSE2 mu an cells (Fig. 7A). gi A ansc ip le els we e high in wild- ype ammonium- g own cells and we e down egula ed upon incuba ion in me- dium con aining ni a e o no combined ni ogen. In con as , gi A ansc ip le els in CSE2 cells emained high unde all condi ions es ed. These esul s demons a e ha N cA e- p esses he gi A p omo e . In addi ion we s udied he accumula ion o he IF7A p o ein in he wild ype and he CSE2 mu an s ain unde he h ee ni ogen egimens analyzed. Fo his pu pose, we pe o med Wes e n blo analysis and ound ha IF7A accumula ed in he wild- ype cells cul i a ed wi h ammonium, bu his p o ein was unde ec able unde he o he ni ogen condi ions. On he o he hand, IF7A was also unde ec able in CSE2 cells unde all condi ions es ed. We also analyzed he le el o GS in bo h s ains unde he same condi ions, using an i-GS an ibodies. As shown in Fig. 7B, he amoun o GS is lowe in CSE2 cells han in he WT, unde he di e en ni ogen egimens. I he GS-IF in e ac ion is c i ical o IF s abili y as desc ibed o he Synechocys is sys em (6), he low le el o he IF7A a ge p o- ein (GS) ound in CSE2 cells may con ibu e o he lack o IF7A accumula ion in his s ain. Howe e , he di e ence in FIG. 4. In i o econs i u ion o Synechocys is and Anabaena GS inac i a ion. Synechocys is GS (1.7 ␮g) and Anabaena GS (2.2 ␮g) we e incuba ed wi h inc easing quan i ies o IF7 (A), IF17 (B), and IF7A (C) in a inal olume o 20 ␮l. Inac i e GS-IF complexes we e allowed o o m du ing 2 min, and GS ans e ase ac i i y was de e mined. One hund ed pe cen ac i i y co esponds o 0.4 uni o GS. VOL. 192, 2010 GS POSTTRANSCRIPTIONAL REGULATION IN ANABAENA PCC 7120 4705 on July 20, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://jb.asm.o g/Downloaded om amoun s o GS be ween he wo s ains, WT and CSE2, is no ha much. To analyze he possibili y ha he absence o IF7A in CSE2 cells was due o ano he eason, we decided o se- quence he PCR-ampli ied gi A gene om genomic DNA o bo h s ains. A poin mu a ion, gi A49C⬎T, encoding he sub- s i u ion Q17STOP, was ound in ou independen PCR am- pli ica ions o he gi A gene om he CSE2 s ain. The mu a- ion posi ion is indica ed as he nucleo ide dis ance om he i s nucleo ide o he s a codon o he mu a ion si e. No mu a ion was ound in he gi A gene ampli ied om he WT Anabaena s ain. The ea ly e mina ion o IF7A p o ein in he CSE2 s ain mus be he eason why we could no obse e FIG. 5. P ime ex ension analysis o he gi A ansc ip . (A) To al RNA (20 ␮g) om mid-log Anabaena cells g own wi h ammonium (NH 4 ⫹ ) o ni a e (NO 3 ⫺ ) o g own wi h ni a e and incuba ed o 7 h in he absence o any ni ogen sou ce (N 2 ) was used o p ime ex ension analysis, as desc ibed in Ma e ials and Me hods. A sequencing ladde ca ied ou wi h he same p ime as ha used o p ime ex ension is also shown. The ansc ip ional s a si e is ma ked wi h an as e isk on he sequence, and a pu a i e ⫺10 sequence is boxed. (B) Alignmen o Anabaena and Synechocys is gi p omo e s. N cA binding si es a e shaded in g ay, and ⫺10 egions a e boxed. The consensus sequence o he N cA-ac i a ed p omo e is also shown. FIG. 6. Gel e a da ion analysis o he binding o N cA o he gi A p omo e egion. Two DNA agmen s we e used: Pgi A1, encompassing he wild- ype gi A p omo e (A), con aining wo pu a i e N cA binding si es, and Pgi A2, deple ed o he i s iple o one o he N cA binding si es (B). Each iple o pu a i e N cA binding si es is ep esen ed by a small black box in Pgi A1 and Pgi A2 schemes. Bo h p obes we e end labeled and incuba ed in he p esence o pu i ied N cA p o ein, 5 o 250 nM (lanes 1 o 7 and 9 o 15, espec i ely). Pgi A1 was also incuba ed wi h 250 nM N cA in he p esence o 0.6 mM 2-oxoglu a a e (lane 7) o a 200- old excess o he same unlabeled agmen (lane 8). 4706 GALMOZZI ET AL. J. BACTERIOL. on July 20, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://jb.asm.o g/Downloaded om accumula ion o ha p o ein in his s ain. To u he in es i- ga e i his inding is meaning ul, we decided o in oduce wo di e en cons uc s in he CSE2 s ain. One o hem con ained he wild- ype gi A gene and i s p omo e egion, and he o he con ained he same DNA agmen bu wi h wo poin mu a- ions in he gi A coding egion: gi A40C⬎T, encoding he sub- s i u ion Q14STOP, and gi A46C⬎T, encoding he subs i u ion Q16STOP. A e conjuga ion, colonies we e ob ained using CSE2 cells as ecipien and ei he plasmid pWTgi A (wild- ype gi A gene) o plasmid pMTgi A (mu an gi A gene). Howe e , u he g ow h o exconjugan s bea ing he wild- ype e sion o gi A in eg a ed in he nucA-nuiA egion o he ␣megaplasmid was e y poo . Two amoun s om cellula suspensions o ex- conjugan s we e spo ed on pla es (Fig. 7C). Whe eas all ex- conjugan s bea ing he mu an e sion o gi A g ew well, he g ow h o s ains con aining he wild- ype e sion o gi A was negligible. Analysis o an Anabaena s ain exp essing wo inac i a ion ac o s. Synechocys is and The mosynechococcus elonga us bo h display apid inac i a ion o GS ollowing addi ion o ammo- nium compa ed o Anabaena (6) (Fig. 1). Since bo h Synecho- cys is and The mosynechococcus elonga us ha bo wo inac i a - ing ac o s, IF7 and IF17, and wo IF17 homologous p o eins, espec i ely, i migh be inqui ed whe he he slow esponse in Anabaena is ela ed o he lack o any IF17 homologous p o- ein. We also know om p e ious s udies ha IF7 and IF17 display di e en s abili ies and ha he 82- esidue-long amino- e minal pa o IF17 may be esponsible o he di e en s abili ies obse ed (6; unpublished esul s). Thus, we p o- ceeded o gene a e an Anabaena s ain exp essing an IF17-like GS inac i a ion ac o . Fo his pu pose we cons uc ed a chi- me ic gene be ween gi B om Synechocys is and gi A om Anabaena, in o de o exp ess a modi ied e sion o IF7A wi h he 82- esidue-long amino- e minal pa o IF17 used o i s amino e minus, unde he con ol o he gi B p omo e . This cons uc was in oduced in Anabaena by conjuga ion (3) and in eg a ed, h ough homologous ecombina ion, in he nucA- nuiA egion o he ␣megaplasmid. The esul ing Anabaena s ain, ACHI, con ains he unal e ed gi A gene in he ch omo- some and he chime ic gene gi B/gi A in he ␣megaplasmid. Be o e analyzing GS inac i a ion/ eac i a ion p ocesses in he ACHI s ain, we wan ed o es in i o inac i a ion o Anabaena GS by he chime ic p o ein IF17N/IF7A. Fo his pu pose we pu i ied his p o ein exp essed in E. coli and s ud- ied i s capaci y o inac i a e Anabaena GS compa a i ely wi h IF7A. Figu e 8A shows ha he chime ic p o ein is much less e ec i e han IF7A on GS inac i a ion. We p oceeded hen o he in i o analysis o WT and ACHI s ains; we es ed in bo h FIG. 7. Exp ession le els o N cA-con olled genes in he CSE2 mu an . Ammonium-g own wild- ype and CSE2 Anabaena cul u es we e di ided in o h ee aliquo s. Aliquo s we e ans e ed o 6h oni a e-con aining medium (NO 3 ⫺ ) o o ni ogen- ee medium (N 2 ) o we e main ained in ammonium-con aining medium (NH 4 ⫹ ). (A) Samples we e aken o o al RNA isola ion. Fi een mic og ams o o al RNA was dena u ed, sepa a ed by elec opho esis, blo ed, and hyb idized wi h gi A and glnA p obes. Hyb idiza ion signals we e quan i ied wi h a Cyclone S o age Phospho sys em au o adiog aphy appa a us. gi A and glnA le els we e no malized o hose o npB. I should be no ed ha 100% co esponds o he maximal signal o hyb idiza ion o each p obe and, hus, signals om di e en p obes canno be compa ed. (B) F om he same cul u es as hose used o No he n blo analysis, samples we e aken o Wes e n blo ing. To al p o eins we e isola ed and esol ed by SDS-PAGE, blo ed, and incuba ed wi h an i-IF7A and an i-T xA an ibodies. The il e was s ipped and incuba ed wi h an i-GS an ibodies. (C) G ow h o CSE2 exconjugan s ha bo ing wild- ype (WT1 o WT4) o mu an (MT1 o MT4) e sions o he gi A gene. Fi e mic oli e s o cellula suspensions om he di e en s ains and a 1/10 dilu ion we e spo ed on ammonium-con aining BG11 pla es. Pho os we e ob ained a e 2 weeks o incuba ion. VOL. 192, 2010 GS POSTTRANSCRIPTIONAL REGULATION IN ANABAENA PCC 7120 4707 on July 20, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://jb.asm.o g/Downloaded om s ains he mRNA le el o gi genes in pa allel wi h IF7A and IF17N/IF7A accumula ion in he ammonium-media ed GS in- ac i a ion/ eac i a ion p ocesses. Figu e 8B shows ha ammo- nium addi ion o ni a e-g own cells p o okes an inc ease o gi A mRNA in he WT s ain and o he gi B/gi A chime ic gene in he ACHI s ain. Wi h espec o he p o ein le els, in he ACHI s ain bo h p o eins, IF7A and IF17N/IF7A, accumu- la ed a e ammonium addi ion and dec eased upon ammo- nium emo al (Fig. 8C). In pa allel wi h No he n and Wes e n blo ing expe imen s, we measu ed GS ac i i y o e he same ime cou se in he wo s ains. A clea di e ence in GS ac i i y could be obse ed a e ammonium addi ion (Fig. 8D). Whe eas he ACHI s ain eached abou 40% o he ini ial ac i i y in he i s 2 h, he WT s ain showed he ypical slow Anabaena GS inac i a ion. Pgi A is ep essed in he e ocys s. The GS egula o y sys em is s ic ly dependen on he global ni ogen con ol egula o N cA in bo h Synechocys is (9) and Anabaena (see abo e). No ably, a di e en ial n cA gene exp ession be ween he e o- cys s and ege a i e cells has been desc ibed in Anabaena (23, 24). The e o e, i would be o in e es o analyze gi exp ession along he Anabaena ilamen s. Fo his s udy we analyzed in i o he exp ession o a gi A-g p ansla ional usion. This con- s uc was in oduced in Anabaena by conjuga ion (3) and FIG. 8. GS inac i a ion/ eac i a ion p ocesses in he ACHI s ain. (A) In i o inac i a ion o Anabaena GS wi h IF7A o IF17N/IF7A. Anabaena GS (1.65 ␮g) was incuba ed wi h inc easing quan i ies o IF7A o IF17N/IF7A in a inal olume o 20 ␮l. Inac i e GS-IF complexes we e allowed o o m o 2 min, and GS ans e ase ac i i y was de e mined. One hund ed pe cen ac i i y co esponds o 0.3 uni o GS. Wild- ype and ACHI Anabaena s ains we e g own in BG11C using ni a e as ni ogen sou ce. A he ime indica ed by an a ow in panel D, 10 mM NH 4 Cl was added. An a ow in panel D also indica es he ime a which cells we e washed wi h ammonium- ee medium and GS eac i a ion ook place. (B) No he n blo assay o he gi genes unde di e en ni ogen condi ions. To al RNA was isola ed om cells g own wi h ni a e (0) and a e ammonium addi ion (⫹NH 4 ⫹ ) a he indica ed imes (min). Gels we e blo ed and hyb idized wi h gi A (WT s ain) and gi B (ACHI s ain) p obes. Fil e s we e s ipped and ehyb idized wi h an npB gene p obe as a loading con ol. (C) Wes e n blo assay o IF7A and IF17N/IF7A p o eins du ing he GS inac i a ion/ eac i a ion p ocesses. Samples we e aken om ni a e-g own cells (0) and a e ammonium addi ion (⫹NH 4 ⫹ )o emo al (⫺NH 4 ⫹ ) a he indica ed imes (min). To al p o eins we e isola ed and esol ed by SDS-PAGE, blo ed, and incuba ed wi h an i-IF7A and an i-T xA an ibodies. (D) F om he same cul u es as hose used o No he n and Wes e n analysis, GS ans e ase ac i i y was de e mined in si u. 4708 GALMOZZI ET AL. J. BACTERIOL. on July 20, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://jb.asm.o g/Downloaded om in eg a ed h ough homologous ecombina ion in he gi A lo- cus. The esul ing Anabaena s ain, AGFP1, was examined by luo escence mic oscopy. A delayed ammonium-dependen in- duc ion o gi genes in ni ogen-s a ed cells has been epo ed o Synechocys is (9). This delay is likely due o he high 2-oxo- glu a a e le els unde hese condi ions (20). Assuming a sim- ila scena io in ni ogen- ixing Anabaena cells (16, 21, 30), we analyzed by No he n blo ing gi A induc ion a e ammonium addi ion in wild- ype Anabaena cells om diazo ophic o ni- a e-supplied cul u es. A delayed gi A induc ion was obse ed in ni ogen- ixing cells (no shown). Taking his in o accoun , we moni o ed Pgi A induc ion in he AGFP1 s ain. Cells o his s ain we e cul i a ed in ni a e-con aining medium and hen ans e ed o ni ogen- ee medium. Once ma u e he - e ocys s we e obse ed by ligh mic oscopy (24 h), 10 mM ammonium was added and samples we e aken o luo es- cence mic oscopy analysis a 3, 4.5, and 6.5 h. Upon ammo- nium addi ion, GFP exp ession in ege a i e cells becomes highe han ha obse ed in he e ocys s. Such di e en ial ex- p ession is clea ly obse ed a 4.5 and 6.5 h (Fig. 9). This obse a ion sugges s ha de ep ession o he gi A p omo e is no obse ed in he e ocys s, and hus, he GS inac i a ion sys em is ep essed in his ype o cell. DISCUSSION The wo k p esen ed he e e eals ha he GS pos ansc ip- ional egula ion sys em desc ibed i s in he Synechocys is sp. PCC 6803 s ain is no es ic ed o his cyanobac e ium. In ac , genes homologous o gi A and gi B om Synechocys is ha e been ound in se e al cyanobac e ial genomes bu seem o be absen in s ains o he genus P ochlo ococcus. He e we show ha he gi A gene om he ilamen ous, ni ogen- ixing cyanobac e ium Anabaena sp. PCC 7120 is esponsible o GS inac i a ion in his o ganism. I is wo h no ing ha he gene ic con ex s o gi A genes in se e al genomes o ilamen ous cya- nobac e ia a e simila . In hese s ains, Anabaena sp. PCC 7120, Anabaena a iabilis ATCC 29413, Nos oc punc i o me sp. PCC 73102, Nodula ia sp. s ain PCC 9350, and Anabaena azollae, he gi A gene is loca ed downs eam and on he oppo- si e s and om he GS-encoding gene, glnA. This ac aises he possibili y o addi ional GS egula o y mechanisms, medi- a ed by he gi A gene, a ec ing glnA a he mRNA le el, which would be conse ed in ilamen ous cyanobac e ia. Fu he - mo e, his p oximal localiza ion o he GS/IF coding genes may be ela ed o genome eo ganiza ion phenomena o coe olu- iona y p ocesses. In his sense, ou obse a ion ha only IF7A is able o inac i a e Anabaena GS in i o (Fig. 4) is consis en wi h his possible coe olu ion. The cha ac e iza ion o gi A exp ession in Anabaena e eals ha ammonium-media ed up egula ion o his gene is no ansi o y, as desc ibed o gi genes in Synechocys is (8). The slow GS inac i a ion obse ed in Anabaena migh be he ea- son why gi A exp ession emains high du ing ammonium ea - men . In Synechocys is, ammonium addi ion p o okes a s ong dec ease in he 2-oxoglu a a e pool and hus gi genes a e de ep essed. Howe e , a quick GS inac i a ion leads o he inc ease o he 2-oxoglu a a e amoun and N cA-media ed e- p ession o gi genes akes place again (9, 20). In his ega d, Synechocys is mu an s ains ha ha bo only he gi A gene beha e simila ly o Anabaena wi h espec o GS inac i a ion and gi A exp ession (no shown). Analysis o he Anabaena gi A p omo e e ealed he p es- ence o wo N cA binding si es; one o hem is cen e ed a posi ion ⫺28.5 in espec o he TSP, which is a localiza ion desc ibed o N cA- ep essed p omo e s, cen e ed down- s eam o posi ion ⫺40.5 (9). The o he si e, cen e ed a posi- ion ⫺77.5, is a pu a i e ac i a o si e because some N cA- ac i a ed p omo e s ha e been desc ibed o bea N cA binding si es ups eam o posi ion ⫺41.5 no ma ching he s uc u e o he canonical N cA-ac i a ed p omo e , wi h an N cA binding box cen e ed a ⫺40.5 (9, 14). One p omo e in which N cA ac s bo h as an ac i a o and as a ep esso has been desc ibed; his is he case o he gl X gene om Synechococcus elonga us, bu i s egula o y pa e n is unique among he N cA- egula ed genes. In he case o he Anabaena gi A gene, he esul s ob- ained wi h he n cA mu an s ain (CSE2) clea ly demons a e FIG. 9. GFP luo escence o Anabaena s ain ca ying he gi A-g p usion (AGFP1). GFP luo escence mic og aphs o diazo ophically g own ilamen s a e incuba ion wi h 10 mM ammonium o 3, 4.5, o 6.5 h. Ligh ansmission mic og aphs (le column), phycobilip o ein au o luo- escence (middle column), and GFP luo escence ( igh column) a e shown o each condi ion. Whi e iangles poin o he e ocys s. VOL. 192, 2010 GS POSTTRANSCRIPTIONAL REGULATION IN ANABAENA PCC 7120 4709 on July 20, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://jb.asm.o g/Downloaded om