N-terminal chimaeras with signal sequences enhance the functional expression and alter the subcellular localization of heterologous membrane-bound inorganic pyrophosphatases in yeast

Ti le: N- e minal chime as wi h signal sequences enhance he unc ional exp ession and al e he subcellula localiza ion o he e ologous memb ane-bound ino ganic py ophospha ases in yeas Running Ti le: Enhanced exp ession o H+-PPases in yeas Au ho s: Rocío DRAKE, Au elio SERRANO1 and José R. PÉREZ-CASTIÑEIRA Ins i u o de Bioquímica Vege al y Fo osín esis, Uni e sidad de Se illa-CSIC, A da. Ame ico Vespucio, 49, 41092 Se illa, Spain 1 To whom co espondence should be add essed (Phone: +-34-954489525. Fax: +-34- 954460065; Email: [email p o ec ed]) Abb e ia ions used: ACMA, 9-amino-6-chlo o-2-me hoxy ac idine; CCCP, cyanide m- chlo ophenylhyd azone; DAPI, 4',6-diamidino-2-phenylindole; FM4-64, N-(3- ie hylammoniump opyl)-4-(p-die hylaminophenyl-hexa ienyl) py idinium dib omide; GFP, g een luo escen p o ein; H+-PPase, p o on- ansloca ing ino ganic py ophospha ase; ORF, open eading ame; PPi, ino ganic py ophospha e; sPPase, soluble ino ganic py ophospha ase; PCR, polyme ase chain eac ion; yEGFP, yeas enhanced g een luo escen p o ein. Biochem J. 2010 Feb 9;426(2):147-57. doi: 10.1042/BJ20091491 Abs ac Exp ession o he e ologous mul ispanning memb ane p o eins in Saccha omyces ce e isiae is a di icul ask. Qui e o en, he use o mul icopy plasmids whe e he o eign gene is unde he con ol o a s ong p omo e does no gua an ee e icien p oduc ion o he co esponding p o ein. He e, we show ha he exp ession le el and/o subcellula localiza ion in S. ce e isiae o an he e ologous ype o mul ispanning memb ane p o ein, he H+- ansloca ing ino ganic py ophospha ase (H+-PPase), can be changed by using i wi h a ious sui able N- e minal signal sequences. Chime ic p o eins we e cons uc ed by adding he pu a i e N- e minal ex a domain o T ypanosoma c uzi H+-PPase o he bona ide signal sequence o S. ce e isiae in e ase Suc2p o H+-PPase polypep ides o di e en o ganisms ( om bac e ia o plan s) and exp essed in a yeas condi ional mu an de icien in i s cy osolic PPi hyd olysis ac i i y when g own on glucose. Chime ic cons uc s no only subs an ially enhanced H+- PPase exp ession le els in ans o med mu an cells bu also allowed unc ional complemen a ion in hose cases in which na i e H+-PPase ailed o accomplish i . Ac i i y assays and Wes e n blo analyses u he demons a ed he occu ence o mos H+-PPase in in e nal memb ane ac ions o hese cells. The addi ion o N- e minal signal sequences o he acuola H+-PPase AVP1 om he plan A abidopsis haliana -a p o ein e icien ly exp essed in yeas in i s na u al o m- al e s he subcellula dis ibu ion o he chime as sugges ing u he p og ession along he sec e o y so ing pa hways, as shown by densi y g adien ul acen i uga ion and in i o luo escence mic oscopy o he co esponding GFP-H+-PPase usion p o eins. Key wo ds: chime ic p o eins, g een luo escen p o ein, he e ologous exp ession, N- e minal signal sequences, p o on- ansloca ing ino ganic py ophospha ase, Saccha omyces ce e isiae. INTRODUCTION He e ologous exp ession may pose impo an di icul ies, especially in he case o mul ispanning memb ane p o eins, whose o e exp ession can be oxic o he hos cell. The use o app op ia e p omo e s o known p ope ies allows he con ol o ansc ip ion; howe e , qui e o en, high le els o messenge RNA do no imply high amoun s o he desi ed p o ein. This sugges s ha , besides ansc ip ion, o he p ocesses such as ansla ion, p o ein a ge ing o deg ada ion may also be c ucial in o de o ob ain high le els o unc ional o eign p o eins in a gi en sys em. Euka yo ic cells ha e an in acellula memb ane ne wo k ha allows hem o accommoda e addi ional memb ane p o eins [1, 2]; he e o e, hey a e equen ly u ilized in o de o exp ess he e ologous mul ispanning memb ane p o eins. In euka yo es, memb ane p o eins a e a ge ed o speci ic cell compa men s o o he plasma memb ane by he endomemb ane sys em, o sec e o y pa hway, ha includes he endoplasmic e iculum (ER), Golgi complex, he acuole and o he anspo esicles, and by he endocy ic pa hway ha e en ually in e nalizes hem o he endosomes [1]. Na i e H+- ATPases ha e been used among o he memb ane p o eins o in es iga e he esicula anspo machine y in he budding yeas Saccha omyces ce e isiae [2, 3], in which he e ologous memb ane p o eins a e usually deli e ed by de aul o he ER. Nascen p o eins in euka yo ic cells mus pass s ingen quali y con ols designed o op imize hei syn hesis, olding and, when necessa y, pos - ansla ional modi ica ions [4, 5]. I is no unusual ha p o eins exp essed in euka yo ic sys ems ail o pass hese con ols being di e ed o p o eoly ic compa men s, esul ing in low o , indeed, no exp ession. In he case o S. ce e isiae, one o he mo e ad an ageous sys ems o he he e ologous exp ession o memb ane p o eins [6], hese p oblems ha e been some imes o e come by u ilizing mu an s de ec i e in hei acuola p o eases [7]; howe e , e en he la e may no be su icien in some cases [8]. Al e na i ely, cons uc ion o chime ic p o eins wi h modi ied N- e minal domains has also been epo ed o imp o e he e ologous exp ession o he in eg al memb ane p o ein SR Ca2+-ATPase in yeas [9]. The la e sugges s ha he N- e minus migh be c ucial o de e mine he a e o memb ane p o eins in his sys em. He e, we show ha he exp ession in S. ce e isiae o se e al bac e ial and euka yo ic memb ane-bound p o on- ansloca ing ino ganic py ophospha ases (H+-PPases), a ype o memb ane-embedded p o eins na u ally absen in yeas , can be imp o ed by a aching di e en signal pep ides o hei N- e mini. Na u ally occu ing H+-PPases a e loca ed in plasma memb ane esicles and acidocalcisomes o p oka yo es and acidic single-memb ane o ganelles (acidocalcisomes, lysosomes, acuoles) o ce ain euka yo es: p o is s and plan s [10-13], al hough a plasma memb ane localiza ion was also epo ed o plan -like K+- s imula ed H+-PPase o some pa asi ic p o is s [14]. In his wo k, chime ic H+-PPase p o eins we e exp essed in YPC3, a no el e sion o a yeas condi ional mu an in i s essen ial cy osolic py ophospha ase (Ipp1p), which is unable o g ow on glucose [15]. In mos cases, exp ession le els o he chime as we e highe han hose o hei na u al coun e pa s; his usually imp o ed he abili y o he di e en H+-PPases o suppo g ow h o YPC3 on glucose by unc ionally complemen ing Ipp1p. In he case o A. haliana acuola H+-PPase AVP1, a p o ein al eady exp essed e icien ly in yeas [15, 16], bo h exp ession le els and memb ane- associa ed PPase ac i i y alues o he chime as we e simila o hose o he na i e p o ein, bu densi y g adien ul acen i uga ion showed al e ed subcellula dis ibu ions, sugges ing di e en p og essions along he sec e o y pa hway. In o de o u he in es iga e his e ec , GFP was used o AVP1, and wo di e en N- e minal signal sequences we e a ached o he esul ing chime ic polypep ide. Fluo escence mic oscopy analysis o yeas cells exp essing hese chime as showed ha N- e minal domains subs an ially modi y he subcellula dis ibu ion o he he e ologously-exp essed H+-PPase p o eins. EXPERIMENTAL Gene a ion o YPC3 s ain YPC3 mu an s ain was ob ained om S. ce e isiae haploid s ain W303-1A (MATa, ade2-1 can1-100 his3-11,15 leu2-3,112 p1-1, u a3-1) by inse ing he yeas HIS3 casse e om plasmid pUG34 (h p://mips.helmhol z-muenchen.de/p oj/yeas /in o/ ools/hegemann/ g p.h ml) ollowed by he galac okinase (GAL1) p omo e be ween he p omo e and he coding sequence o he gene ha codes o he cy osolic ino ganic py ophospha ase (IPP1) using he single-s ep ansplacemen p ocedu e [17]. Fo his pu pose, an app op ia e linea DNA agmen was ob ained as ollows: i s , he coding sequence o IPP1, along wi h a 390 bp-long egion ups eam he pu a i e s a codon [18], was ampli ied om genomic DNA by high ideli y PCR using TaqPlus DNA polyme ase (S a agene) and liga ed in o p-GEM-T plasmid (P omega), ollowing he manu ac u e s' ins uc ions. Then, wo DNA agmen s con aining he yeas HIS3 casse e (1.7 kb long) and he GAL1 p omo e (0.75 kb long), espec i ely, we e sequen ially in oduced a he EcoRV si e loca ed 21 bases ups eam he s a ing codon o IPP1 [18] by s anda d cloning echniques [19]. This yielded plasmid pYPC3 ha was diges ed wi h ApaI and SacI and, hen, in oduced in S. ce e isiae haploid s ain W303-1A, by he LiOAc-PEG me hod [20]. T ans o man s we e selec ed by g owing cells on 2% aga pla es made in galac ose-con aining cul u e medium [21] wi hou his idine. The inse ion o he cons uc ion was checked by PCR pe o med wi h genomic DNA o one o he ans o man s unable o g ow on glucose (YPC3). Mul iple sequences alignmen s and signal pep ide p edic ions Mul iple amino acid sequence alignmen s o selec ed euka yo ic and p oka yo ic H+-PPases we e done using he Clus al X .1.8 p og am [22] and co ec ed manually. Sequence da a om public da abases and un inished genome p ojec s we e ob ained by simila i y sea ches using BLAST algo i hms agains websi es o he Na ional Cen e o Bioin o ma ics (NCBI), USA (h p://www.ncbi.nih.go /PMGi s/Genomes/allo g.h ml), he Join Genomic Ins i u e (JGI), USA (h p://spide .jgi-ps .o g/JGI_mic obial/h ml/), he Sange Ins i u e, UK (h p://www.sange .ac.uk/P ojec s/), o he Ins i u e o Genomic Resea ch (TIGR), USA (h p://www. ig .o g/ db/mdb/mdb.h ml). N- e minal ex ension analyses o di e en H+- PPases o p edic ion o signal pep ides we e pe o med wi h he SignalP 3.0 [23] and SIG- P ed p og ams, a ailable a h p://www.cbs.d u.dk/se ices/SignalP/ and h p://bmbpcu36.leeds.ac.uk/p o _analysis/Signal.h ml, espec i ely. Plasmid cons uc ions Se e al de i a i es o he URA3-con aining E. coli/S. ce e isiae shu le plasmid pRS699 [24] we e ob ained in o de o exp ess di e en p o eins in YPC3. As a i s s ep, an a i icial SpeI si e was in oduced by PCR downs eam he XhoI si e loca ed be ween he PMA1 p omo e and e mina o o pRS699. This allowed di ec ional cloning in he esul ing plasmid (named pRS699 b). The coding sequences o he H+-PPase (TcVP) he pa asi ic p o is (kine oplas id) T ypanosoma c uzi s ain Y (accession numbe AF159881) and he S. ce e isiae gene o sec e ed in e ase p ecu so SUC2 (accession numbe AAFW02000123) we e in oduced in o pRS699b ollowing he same s a egy: high ideli y PCR ampli ica ion ( om cDNA o genomic DNA, espec i ely) wi h p ime s con aining a i icial SalI and SpeI si es, diges ion o he ampli ied agmen s wi h hese enzymes and liga ion o pRS699b clea ed wi h XhoI and SpeI. These cons uc ions we e subsequen ly u ilized as empla es in o de o ob ain he PMA1 p omo e ollowed by he sequences coding o he N- e minal domains con aining he signal sequences o TcVP and Suc2p by PCR (Table S1). An a i icial EcoRI si e a he 5'-end and wo consecu i e SalI and SpeI si es a he 3'-end o each o he ampli ied DNA agmen s we e in oduced, so ha hey could be inse ed in plasmid pRS699b by u ilizing he EcoRI and SpeI si es o he la e . This yielded plasmids pTC and pINV. Plasmids pHisTC and pHisINV we e ob ained by he same p ocedu e excep ha a sequence coding o he epi ope MGRGSH6 was in oduced a he 5’-ends o TgVP and SUC2 in hei ampli ica ion eac ion. Finally, he gene coding o yeas -enhanced GFP (yEGFP) was ob ained by PCR using plasmid pUG34 (see abo e) as a empla e and in oducing a i icial XhoI and SalI si es a he 5'-end and 3'-end, espec i ely. The esul ing DNA agmen was liga ed in he igh sense a he unique XhoI si e o pRS699b and SalI si es o pTC and pINV, espec i ely. This yielded plasmids pGFP, pTcGFP and pIn GFP. All hese plasmids we e sequenced o ule ou possible e o s p oduced by he PCR eac ions. The coding sequences o di e en H+-PPases we e ampli ied by PCR om genomic DNA p epa a ions o cDNA clones using p ime s wi h app op ia e a i icial es ic ion si es and inse ed be ween he unique XhoI/SalI and SpeI si es o plasmids pRS699b, pTC and pINV. AVP1 (accession numbe A38230) was also in oduced in plasmids pHisTc, pHisINV, pGFP, pTcGFP and pIn GFP. Ano he e sion o AVP1 con aining he sequence coding o epi ope MGRGSH6 a he 5'-end, ob ained by PCR, was inse ed in plasmid pRS699b, yielding pHisAVP1. All plasmids u ilized o subsequen expe imen s a e lis ed in Table S2. T ans o ma ion and g ow h o ans o med YPC3 cells S. ce e isiae mu an s ain YPC3 was ans o med wi h he plasmids shown in Table 2 as desc ibed [20]. YPC3 cells we e ini ially g own a 30 °C in galac ose-con aining syn he ic medium de oid o his idine. T ans o man s we e selec ed by g owing cells on 2% aga pla es in cul u e medium wi hou his idine and u acil. Complemen a ion s udies o he cy osolic ino ganic py ophospha ase by he H+-PPases we e pe o med by p e-inocula ing 2 ml o galac ose-con aining selec i e medium wi h ans o med cells om he pla es and g owing o e nigh a 30 ºC wi h agi a ion (200 .p.m.). The ollowing day, 2 ml o glucose-con aining selec i e medium we e inocula ed wi h 20 μl o cells g own on galac ose and allowed o g ow as abo e. This ea men is necessa y o b ing down he py ophospha ase ac i i y associa ed o Ipp1p [15]. A e g ow h on glucose, se ial 10- old dilu ions o he cul u es we e made in s e ile wa e and 10 μl d ops o each dilu ion we e spo ed on o YPD and YPGal aga pla es. Pla es we e g own a 30 ºC o 5-6 days. Isola ion o yeas memb anes and suc ose g adien ac iona ion P epa a ion o memb ane ac ions om YPC3 cells ans o med wi h he di e en plasmids we e ob ained by a modi ica ion o a me hod p e iously desc ibed [24]: yeas colonies we e collec ed om a pla e and liquid-g own up o s a iona y phase in galac ose-con aining selec i e medium; hen, 4 ml o s a iona y cul u es we e used o inocula e 400 ml o YPD. A e o e nigh g ow h, cells we e sedimen ed by cen i uga ion a 700 g o 10 min, washed ho oughly wi h wa e , esuspended in 5 ml o ice-cold bu e A (25 mM T is-HCl, pH 8, 10% glyce ol, 4 mM β-me cap oe hanol, 2 mM DTT, 2 mM EDTA, 10 mM MgCl2, 1 mM benzamidine, 2 mM ε-aminocap oic acid, 1 mM PMSF), and homogenized by igo ous shaking wi h glass beads. The homogena e was dilu ed up o 25 ml wi h bu e B (10 mM T is-HCl, pH 7.6, 10% glyce ol, 2 mM DTT, 1 mM EDTA) and cen i uged o 10 min a 700 g o emo e beads and deb is. The esul ing supe na an was cen i uged o 20 min a 20,000 g. The “high densi y memb anes” (HDM) pelle hus ob ained was homogenized in s ipping bu e [25] and sedimen ed (20 min a 20,000 g), whe eas he supe na an was u he cen i uged o 30 minu es a 120,000 g hus yielding a “low densi y memb anes” (LDM) pelle . This pelle was also aken up and homogenized in s ipping bu e and cen i uged o 20 minu es a 120,000 g. Finally, bo h HDM and LDM pelle s we e washed sepa a ely wi h bu e B and cen i uged (20 min, 120,000 g). The inal pelle s we e esuspended and homogenized in 1 ml o bu e B. HDM p epa a ions ob ained om mu an YPC3 exp essing he di e en e sions o AVP1 we e applied on o con inuous 20-50% (w/w) suc ose g adien s made by di usion [24]. A e an o e nigh cen i uga ion (12 h.) a 100,000 g, ac ions o app oxima ely 0.6 ml we e collec ed and dis ibu ed in mic o uge ubes using a pe is al ic pump. Py ophospha ase ac i i y and H+- ansloca ion assays, Wes e n blo and o al p o ein es ima ion Py ophospha ase ac i i y and p o on- ansloca ion assays, and Wes e n blo s we e pe o med as desc ibed elsewhe e [26-28]. P o on- ansloca ion was assayed by eco ding he quenching o 9-amino-6-chlo o-2-me hoxy ac idine (ACMA). This has been shown o be he p obe o choice o he measu emen o ela i ely small ∆pH g adien s [29]. Fo immunode ec ion o H+-PPases, wo p ima y an ibodies we e u ilized: one, aised agains he syn he ic pep ide HKAAVIGDTIGDPLKC, a consensus H+-PPase sequence [16], was ob ained om SIGMA (S . Louis, MI, USA) and ou inely u ilized a 1:1000 dillu ion. The o he was a polyclonal an ibody aised agains he pu i ied H+-PPase om he ma ine bac e ium The mo oga ma i ima [30]. This an ibody was a ini y-pu i ied as desc ibed elsewhe e [31] and u ilized a 1:100 dillu ion. Fo s udies abou he p ocessing o N- e minal ends o chime ic p o eins, a monoclonal an ibody agains he epi ope RGSH4 (QIAGEN, Ge man own, MD, USA) was u ilized. Fo localiza ion o di e en memb ane ac ions in g adien s he ollowing an ibodies agains ma ke p o eins we e used: an i-Pma1p (polyclonal, gene ously p o ided by P o . R. Se ano), an i-Vma1p, an i-Vph1p and an i-Dpm1p (monoclonals, pu chased om Molecula P obes, Eugene, OR, USA). To al p o ein was es ima ed by he B ad o d me hod [32] wi h o oalbumin as a s anda d. In si u luo escence mic oscopy me hods YPC3 cells ans o med wi h plasmids pGFPAVP1, pTcGFPAVP1 and pIn GFPAVP1 and g own in YPD we e ixed wi h 70% e hanol and s ained wi h DAPI as desc ibed elsewhe e [33]. FM4-64 (Molecula P obes), a i al lipophilic s ain ha shows ed luo escence when bound o memb anes o acidic in e nal compa men s, was used wi h Texas Red il e s o co- localiza ion s udies wi h GFP- agged H+-PPase [34]. Images we e aken wi h a ully au oma ed Leica DM6000B mic oscope (Leica Mic osys ems, We zla , Ge many) wi h a 100x objec i e and equipped wi h a cooled CCD came a (ORCA-AG, Hamama su Pho onics, Hamama su, Japan) wi h a Z-op ical spacing o 0.2 µm. RESULTS The H+-PPase om he kine oplas id p o is T ypanosoma c uzi e icien ly complemen s Ipp1p and has a pu a i e N- e minal signal pep ide sequence A yeas mu an bea ing an inse ion o he HIS3 casse e and he GAL1 p omo e be ween he p omo e and he coding sequence o gene IPP1 was gene a ed. The ch omosomal inse ion was checked by PCR, hus, a agmen o app oxima ely 3.9 kb was ampli ied using genomic DNA o a selec ed ans o man (YPC3) as empla e whe eas a 1.3 kb long agmen was ob ained in he case o s ain W303-1A (Fig. 1). Analysis by No he n blo and ac i i y assays pe o med wi h YPC3 cells g own up o s a iona y phase on galac ose and, hen, ans e ed o glucose o u he 12 hou s showed ha IPP1 was ansc ibed only in he p esence o galac ose (no shown), he eby demons a ing ha he only unc ional copy o his gene in YPC3 was unde he con ol o he GAL1 p omo e . T ans o ma ion o YPC3 wi h a plasmid bea ing he coding sequence o T. c uzi H+- PPase (TcVP) unde he con ol o cons i u i e PMA1 p omo e allowed he mu an o eco e he capaci y o g ow e icien ly on glucose (Fig. 2A). Ac i i y assays u he showed K+- s imula ed PPase ac i i y in memb ane p epa a ions o he o me cells and no ac i i y in cy osolic ac ions (no shown). A de ailed compa a i e analysis o he deduced amino acid sequence o his p o ein e ealed an ex a N- e minal domain wi h espec o he majo i y o H+-PPases om o he sou ces, such as hose well s udied ones om he plan A abidopsis haliana o he p o eobac e ium Rhodospi illum ub um (RVP). The H+-PPases om o he pa asi ic p o is s like he kine oplas ids T ypanosoma b ucei (TbVP) and Leishmania majo (LmVP) o he apicomplexan Toxoplasma gondii (TgVP1) also showed ex a N- e minal domains o di e se leng h (Fig. 2B); howe e , none o hese p o eins suppo ed YPC3 g ow h on glucose (no shown). In e es ingly, whe eas he h ee analyzed kine oplas id sequences exhibi de ini e N- e minal signal sequences in he ange o 20 o 24 amino acids including a pu a i e ansmemb ane egion as p edic ed by he SignalP and SIG-P ed me hods, he T. gondii H+-PPase p ecu so holds a longe N- e minal ex ension wi h a no so well de ined bipa i e s uc u e (see Fig. 2B). No signal sequences we e p edic ed o he o he H+-PPases om bo h p oka yo es and euka yo es o ganisms analyzed. I should be no ed ha he H+- PPase o T. c uzi has been p e iously iden i ied by immunochemical me hods in he Golgi appa a us and, o a lesse ex en , in plasma memb ane esicles o his pa asi ic p o is [14]. Addi ion o signal sequences a he N- e minus o se e al H+-PPases inc eases hei exp ession le els and/o hei capaci y o unc ionally complemen Ipp1p Cons uc s composed o he 5’-end egion o TcVP (i s p edic ed 24 amino acids long signal pep ide, plus 4 esidues a e he pu a i e clea age si e) (see Table S1) ollowed by he coding sequences o di e en genes o H+-PPases we e ob ained and inse ed in he cloning si e o a mul icopy shu le plasmid unde he con ol o a cons i u i e p omo e . Simila cons uc ions made wi h he 5´- egion o yeas gene SUC2, encoding he 19 amino acids long signal sequence o sec e ed in e ase (see Table S1), we e also es ed o compa ison. YPC3 was ans o med wi h he di e en plasmids hus ob ained (Table S2) and g ow h o he ans o man s on glucose was checked. Two main scena ios we e obse ed: (1) se e al H+- PPases could no unc ionally subs i u e o Ipp1p o iginally bu gained his abili y when an app op ia e N- e minal signal domain was added, and (2) A. haliana AVP1 could complemen Ipp1p in i s o iginal o m bu his capaci y was clea ly enhanced when used wi h any o he N- e minal domains es ed. In one case, ha o LmVP, he esul ing chime a showed exp ession le els signi ican ly highe han he na u al p o ein; howe e , i could no unc ionally subs i u e o Ipp1p (no shown). Figu e 3 shows he esul s ob ained wi h he K+-independen H+-PPase om he pho osyn he ic p o eobac e ium R. ub um (RVP). T ans o ma ion o YPC3 wi h plasmids pRVP and pTcRVP did no allow he mu an o g ow on glucose; howe e , cells ans o med wi h plasmid pIn RVP e icien ly g ew on his ca bon sou ce. Assays o PPase ac i i y and Wes e n blo analysis u he showed he p esence o signi ican amoun s o a K+-independen H+-PPase wi h an appa en molecula mass (ca. 60 kDa) smalle han ha o o he H+-PPases in memb ane ac ions o he la e cells. In e es ingly, In RVP was by a he mos e ec i ely memb ane a ge ed p o ein (see Fig. 3C). Simila esul s we e ob ained wi h ano he p oka yo ic K+-independen H+-PPase, om he ac inobac e ium S ep omyces coelicolo (no shown). In he case o he K+-dependen H+-PPase TgVP1 o he apicomplexan p o is T. gondii, only YPC3 cells ans o med wi h plasmid pTcTgVP, encoding a usion p o ein wi h he pu a i e signal sequence o T. c uzi H +-PPase, could eco e he capaci y o g ow on glucose. PPi hyd olysis ac i i y assays and Wes e n blo s consis en ly demons a ed he p esence o a K+-dependen PPase in memb anes o hese cells; howe e , yeas cells ans o med wi h plasmid pIn TgVP, encoding a usion p o ein wi h he signal sequence o yeas Suc2p, also exp essed he p o ein (Fig. S1). YPC3 cells ans o med wi h plasmid pTcAVP1 could g ow on glucose mo e e icien ly han hose ans o med wi h plasmids pAVP1 o pIn AVP1; howe e , ac i i y assays and Wes e n blo analysis showed no signi ican di e ences in memb ane-associa ed ac i i y o p o ein le els (Fig. 4). In all cases, H+- ansloca ion ac i i ies o memb ane p epa a ions we e p opo ional o he PPi hyd olysis ac i i ies es ed (Figs. 3D, S1D and 4D). The N- e minal signal sequences om TcVP and Suc2p a e p ocessed when he chime as wi h AVP1 a e exp essed in YPC3 In o de o check whe he he N- e minal signal sequences o TcVP and Suc2p we e e ec i ely p ocessed du ing deli e y and memb ane so ing o he chime ic p o eins, YPC3 cells we e ans o med wi h plasmids pHisAVP1, pHisTcAVP1 and pHisIn AVP1 (Table S2). Immunoblo s o memb ane p epa a ions ob ained om he di e en ans o man s using an an ibody agains he epi ope RGSH4 ecognised a polypep ide only in he case o cells ans o med wi h pHisAVP1; howe e , he an ibody agains a highly conse ed H+-PPase sequence iden i ied polypep ides o he igh size (ca. 70 kDa) in all cases (Fig. S2). Ac i i y assays da a we e simila o hose shown in Fig. 4. These esul s clea ly indica e ha he N- e minal domain o TcVP holds a bona ide signal sequence ha , like he homologous signal pep ide o Suc2p, is co ec ly p ocessed in yeas . To ou knowledge, his is he i s di ec e idence epo ed o p o ein p ocessing and so ing o a H+-PPase, in an he e ologous sys em in his case. An N- e minal usion o TcVP wi h he RGSH4 domain ha is unc ionally exp essed in YPC3 seems o be p ocessed as well, bu un o una ely he chime ic p o ein is no ecognized by he H+-PPase an ibodies used in his s udy (da a no shown). Na u al AVP1 and i s de i ed N- e minal chime as TcAVP1 and In AVP1 exhibi di e en subcellula dis ibu ions in yeas cells F ac iona ion by suc ose g adien ul acen i uga ion o high densi y memb anes ob ained om yeas cells ans o med wi h plasmids pAVP1, encoding na u al plan acuola AVP1, and pTcAVP1 and pIn AVP1, encoding de i ed N- e minal chime as wi h signal sequences, showed signi ican di e ences in he subcellula dis ibu ions o H+-PPase p o ein and PPi hyd oly ic ac i i y. In he case o pAVP1, ac i i y was p e e en ially associa ed o in e nal memb anes (Golgi complex/ acuoles), wi h some ac i i y associa ed o he plasma memb ane. In cells ans o med wi h pIn AVP1, an analogous dis ibu ion was ound, al hough plasma memb ane-associa ed ac i i y was clea ly highe . In con as o ha , memb ane p epa a ions om cells ans o med wi h TcAVP1 showed i ually all he PPase ac i i y o be associa ed o ER/Golgi/ acuola memb anes (Fig. 5). Wes e n blo analyses showed AVP1 o be e enly dis ibu ed be ween g adien ac ions 3 and 12 (being ac ion 1 he bo om o he g adien ), whe eas TcAVP1 appea ed p e e en ially in ac ions 8-12 (Golgi complex/ acuole). S ikingly, in he case o In AVP1 mos o he p o ein appea ed associa ed o he plasma memb ane ( ac ions 3-5) (Fig. 5). In si u luo escence mic oscopy con i ms subcellula dis ibu ion esul s ob ained wi h suc ose densi y g adien s The ques ion o he subcellula dis ibu ion o he di e en e sions o AVP1 was u he in es iga ed by luo escence mic oscopy o YPC3 cells. T ans o man s wi h plasmids pGFPAVP1 and pIn GFPAVP1, encoding espec i ely a GFP-AVP1 usion p o ein wi hou and wi h he Suc2p signal sequence, showed a e y cha ac e is ic pa e n o g een luo escence a ound he nucleus (cis e nae esicles o he ER) and in he plasma memb ane. On he con a y, g een luo escence in cells ans o med wi h plasmid pTcGFPAVP1 appea ed exclusi ely associa ed o he in e nal memb ane ne wo k, showing a sp ead spo ed dis ibu ion h oughou he YPC3 cells (Fig. 6A). Labelling o he di e en ans o med yeas cells wi h FM4-64 clea ly shows ha GFP-AVP1 and In GFP-AVP1 do no co-localize wi h he acuola memb ane; by con as , TcGFP-AVP1 seems o pa ially co-localize wi h a high numbe o small acidic esicles (Fig. 6B). DISCUSSION A S. ce e isiae haploid s ain (YPC3), whose gene coding o he essen ial cy osolic soluble PPase (IPP1) is unde he con ol o he glucose- ep essable galac okinase (GAL1) p omo e , has been gene a ed by he single-s ep ansplacemen p ocedu e. This app oach esul ed in he inse ion o he yeas HIS3 ma ke gene ollowed by he GAL1 p omo e be ween he p omo e and he coding sequence o IPP1. Consequen ly, he condi ional mu an s ain gene a ed YPC3 canno g ow in he p esence o glucose because Ipp1p, essen ial o g ow h, is no exp essed unde hese condi ions. YPC3 is mo e con enien han YPC1, a mu an wi h he same pheno ype p e iously desc ibed [15], o check unc ional complemen a ion o Ipp1p due o i s highe genomic s abili y. The H+-PPase o he kine oplas id p o is T. c uzi (TcVP) has a pu a i e N- e minal ex ension wi h he hallma ks o ype I memb ane p o ein signal sequences, as de e mined by in silico s udies: a hyd ophobic co e, which p esumably comp ises a ansmemb ane (TM) segmen , wi h wo lanking g oups o cha ged esidues and a pu a i e pep idase clea age si e a he C- e minus o he signal pep ide [35]. The obse a ion ha he g ow h o YPC3 on glucose was e icien ly suppo ed by TcVP p omp ed us o in es iga e he possible e ec s ha his domain (absen in many o he p o eins o his class) migh ha e on he he e ologous exp ession in yeas o o he H+-PPases a ailable in ou labo a o y. Compa a i e s udies o hese e ec s wi h hose ob ained wi h chime as made wi h he N- e minal domain o a S. ce e isiae p o ein we e also pe o med. Suc2p N- e minal signal pep ide was chosen o wo main easons: (a) he ini ial s eps o mul ispanning memb ane p o eins ansla ion a e analogous o hose o sec e ed p o eins syn hesis [36], (b) his signal pep ide had been s udied in signi ican de ail and u ilized o e icien ly ansloca e polypep ides ac oss he ER memb ane [37-39]. AUTHOR CONTRIBUTION R. D ake ca ied ou mos o he expe imen al wo k, excep gene a ion o YPC3 and luo escence mic oscopy. J. R. Pé ez-Cas iñei a supe ised R. D ake’s wo k, designed he s udy, supe ised da a collec ion, gene a ed YPC3 mu an , pe o med mic oscopy wo k and d a ed he manusc ip . A. Se ano ini ially concei ed he s udy, did he in silico analyses, co- supe ised he expe imen al wo k and da a collec ion and con ibu ed o he manusc ip d a . ACKNOWLEDGEMENTS We g a e ully acknowledge D . Rosa L. López-Ma qués (Uni . o Copenhagen, Denma k), P o . And és Aguile a (Uni . o Se ille, Spain), P o s. Sil ia Mo eno and Robe o Docampo (Uni . o Geo gia, A hens, GA, USA) o gene ously p o iding an ibody agains T. ma i ima H+-PPase, plasmid pUG34 and he cDNA clones o T. gondii and T. c uzi H +-PPases, espec i ely. We also hank D . Alicia O ea o excellen echnical assis ance wi h luo escence mic oscopy. FUNDING This wo k was suppo ed by esea ch g an s om he Spanish Minis e io de Ciencia e Inno ación (BMC2007-61887) and Regional Andalusian Go e men (PAIDI g oup BIO- 261). A pa o unds came om he FEDER p og am o he EU. LEGENDS OF FIGURES Figu e 1. Gene a ion o YPC3, a new yeas mu an whose essen ial cy osolic ino ganic py ophospha ase Ipp1p is unde he con ol o he galac okinase (GAL1) p omo e . (A) The gene coding o he cy osolic py ophospha ase o yeas haploid s ain W303-1A was dis up ed by inse ion o a linea DNA agmen comp ising he yeas HIS3 casse e as a selec ion ma ke ollowed by he galac okinase (GAL1) gene p omo e . A ows ep esen he oligonucleo ides u ilized as PCR p ime s. (B) Analysis by PCR o he inse ion, genomic DNA ob ained om s ains YPC3 and W303-1A we e u ilized as empla es and oligonucleo ides co esponding o he p omo e egion and he end o he coding sequence o IPP1 as p ime s. (C) Pheno ypic cha ac e iza ion o YPC3 by checking i s g ow h on glucose as compa ed wi h he yeas s ain u ilized o i s gene a ion. See main ex o de ails. Figu e 2. Func ional complemen a ion o he cy osolic py ophospha ase Ipp1p o S. ce e isiae by he H+-PPase om he kine oplas id p o is T ypanosoma c uzi (A) and mul iple sequences alignmen o he N- e minal egions o H+-PPase o hologs om selec ed euka yo ic (bold) and p oka yo ic (plain) o ganisms (B). N- e minal ex ensions ha ing he gene al ea u es o signal pep ides, namely, a hyd ophobic co e and a posi i ely cha ged N- e minal domain, and p edic ed as such by he SignalP and SIG-P ed p og ams a e shown boxed and he co esponding pu a i e clea age si es indica ed by e ical a owheads. Posi i ely cha ged (K,R) and hyd ophobic (A,V,L,I,W,F,M) amino acid esidues in he pu a i e signal sequences a e ep esen ed in i alics and in bold ace wi h g ey backg ound, espec i ely. No e he bipa i e signal pep ide wi h wo pu a i e clea age si es p edic ed o he H+-PPase o he apicomplexan p o is Toxoplasma gondii, o which he N- e minal esidues o unca ed cons uc s used by D ozdowicz e al. [8] o exp ession in yeas a e indica ed by olded a ows. App oxima e loca ions o he ansmemb ane segmen (TMS) cha ac e is ic o signal leade pep ides and he i s one (TMS1) in all H+- PPases a e also shown [48]. The p o ein sequences used a e (accession numbe s in pa en heses): T ypanosoma c uzi (AAF80381); T ypanosoma b ucei (AAK95376); Leshmania majo (CAJ08309); T. gondii TgVP1 (AAK38077); Plasmodium alcipa um P VP1 (AAD17215); A abidopsis haliana AVP1 (A38230); O yza sa i a (BAA08232); The mo oga ma i ima (D72049); Rhodospi illum ub um (AAC38615); Me hanosa cina mazei (AAM22543); Py obaculum ae ophilum (AAF01029); S ep omyces coelicolo (T36668), and Chlo obium epidum (con ig g12, NC_002932). The unpublished H+-PPase sequences om he mic oalga Chlo ella sp. NC64A we e iden i ied by hei JGI genome p ojec code numbe s. Figu e 3. D op es s (A), memb ane-associa ed PPi hyd olysis ac i i y assays (B), Wes e n blo analysis (C) and aces o ACMA luo escence quenching by memb ane p epa a ions (D) o yeas mu an s ain YPC3 ans o med wi h plasmids pRS699b, pRVP, pTcRVP and pIn RVP. Expe imen s we e pe o med wi h h ee independen clones aken om each o he ans o man pla es. (A) Typical d op es . Ac i i y alues shown in (B) a e means + S.E. (n=3) using LDM p epa a ions ob ained om YPC3 cells ans o med wi h pRS699b as e e ence Whi e and black ba s ep esen ac i i ies in he absence and p esence o po assium chlo ide, espec i ely. (C) Immunode ec ion o low densi y memb ane (LDM) p epa a ions as desc ibed in Expe imen al P ocedu es. App ox. 70 µg o p o ein we e loaded pe lane. (D) Typical ace o ACMA luo escence quenching ob ained wi h LDM p epa a ions o YPC3 ans o med wi h plasmid pIn RVP. No quenching was obse ed in he case o RVP whe eas, in he case o TcRVP, an in e media e ace be ween hose o C- and In RVP was ob ained (no shown). Figu e 4. D op es s (A), memb ane-associa ed PPi hyd olysis ac i i y assays (B), Wes e n blo analysis (C) and aces o ACMA luo escence quenching by memb ane p epa a ions (D) o yeas mu an s ain YPC3 ans o med wi h plasmids pRS699b, pAVP1, pTcAVP1 and pIn AVP1. Same condi ions as in Fig. 3 excep ha simila aces o ha o AVP1 we e ou inely ob ained wi h LDM p epa a ion o YPC3 cells ans o med wi h plasmid pIn AVP1. Figu e 5. Dis ibu ion o memb ane-bound PPi hyd olysis ac i i y on a con inuous 20%-50% (w/w) suc ose g adien . High densi y memb ane p epa a ions ob ained om YPC3 yeas cells ans o med wi h pAVP1 (), pTcAVP1 () and pIn AVP1 () we e loaded on o suc ose g adien s as desc ibed in Expe imen al P ocedu es. A e o e nigh cen i uga ion, 0.6 ml samples we e collec ed using a pe is al ic pump and ac i i y was measu ed. Ac i i y da a a e a e age o h ee expe imen s pe o med wi h independen clones. To al PPase ac i i y eco e ed om he g adien s we e o he same o de o magni ude. Sample 1 co esponds o he bo om o he g adien . Immunoblo s pe o med using samples 3 o 14 om he g adien s and an a ini y- pu i ied an ibody agains he T. ma i ima H +-PPase [30,31] a e shown. G adien samples (ap ox. 100 µl) we e concen a ed by p ecipi a ion wi h 10% TCA be o e loading. An ibodies agains se e al ma ke p o eins we e also u ilized o localize di e en yeas memb anes: Vph1p, acuola memb ane; Dpm1, ER memb ane; Vma1p, acuola and Golgi memb anes; Pma1p, plasma memb ane. Figu e 6. In si u luo escence mic oscopy o YPC3 cells ans o med wi h plasmids pGFPAVP1, pIn GFPAVP1 and pTcGFPAVP1. (A) Open ield and de ails o a single cell a e shown. Panels show clockwise om bo om le : Noma ski, GFP- agged p o eins, DAPI s aining o nuclei and he o e lap o he h ee images. (B) Visualiza ion o GFP- agged p o eins and memb anes o in e nal acidic compa men s wi h ed luo escen i al dye FM4-64. Panel shows clockwise om bo om le : Noma ski, GFP- agged p o eins, FM4-64 s aining and he o e lap. Cells we e p ocessed and isualized as desc ibed in Expe imen al P ocedu es. Images we e decon ol ed using he Leica LAS-AF so wa e. Maximum p ojec ion images a e shown. Supplemen a y in o ma ion sec ion TABLES Table S1. Nucleo ide and deduced amino acid sequences co esponding o he N- e minal signal domains o T. c uzi H+-PPase and S. ce e isiae in e ase Suc2p used o he cons uc ion o chime as, including he p edic ed clea age si es (a owheads) TcVP ATG GGT GAC ATG AAG AGG TTC ATC GTG GCG GTC GCG GCG GTG TGT TTA CTC M G D M K R F I V A V A A V C L L GCA GCC ACC GTG TCG GCC▼GCA CCG GCG GGG GGT GAG A A T V S A A P A G G E Suc2p ATG CTT TTG CAA GCT TTC CTT TTC CTT TTG GCT GGT TTT GCA GCC AAA ATA M L L Q A F L F L L A G F A A K I TCT GCA▼TCA ATG ACA AAC GAA ACT S A S M T N E T Table S2. Plasmids u ilized o YPC3 ans o ma ion. O iginal and chime ic nucleo ide sequences desc ibed in he Table we e inse ed be ween yeas PMA1 p omo e and e mina o o he URA3-con aining plasmid pRS699b (see Expe imen al o de ails) Plasmid Tag N- e minal signal pep ide GFP H+-PPase coding sequences pRVP - - - Rhodospi illum ub um H+-PPase (RVP) pTcRVP - T ypanosoma c uzi H+-PPase (TcVP) (Table S1) - RVP pIn RVP - Saccha omyces ce e isiae Suc2p (Table S1) - RVP pTgVP - - - Toxoplasma gondii H+-PPase I (TgVP1) pTcTgVP - TcVP - TgVP1 pIn TgVP - Suc2p - TgVP1 pAVP1 - - - A abidopsis haliana H+-PPase I (AVP1) pTcAVP1 - TcVP - AVP1 pIn AVP1 - Suc2p - AVP1 pHisAVP1 RGSH6 - - AVP1 pHisTcAVP1 RGSH6 TcVP - AVP1 pHisIn AVP1 RGSH6 Suc2p - AVP1