Late Assembly Steps and Dynamics of the Cyanobacterial Photosystem I
2007; Elsevier BV; Volume: 282; Issue: 15 Linguagem: Inglês
10.1074/jbc.m609206200
ISSN1083-351X
AutoresUlf Dühring, Friedrich Ossenbühl, Annegret Wilde,
Tópico(s)Photoreceptor and optogenetics research
ResumoThe dynamics of photosystem I assembly in cyanobacteria have been addressed using in vivo pulse-chase labeling of Synechocystis sp. PCC 6803 proteins in combination with blue native polyacrylamide gel electrophoresis. The analyses indicate the existence of three different monomeric photosystem I complexes and also the high stability of photosystem I trimers. We show that in addition to a complete photosystem I monomer, containing all 11 subunits, we detected a PsaK-less monomer and a short-lived PsaL/PsaK-less complex. The latter two monomers were missing in the ycf37 mutant of Synechocystis sp. PCC 6803 that accumulates also less trimers. Pulse-chase experiments suggest that the three monomeric complexes have different functions in the biogenesis of the trimer. Based on these findings we propose a model where PsaK is incorporated in the latest step of photosystem I assembly. The PsaK-less photosystem I monomer may represent an intermediate complex that is important for the exchange of the two PsaK variants during high light acclimation. Implications of the presented data with respect to Ycf37 function are discussed. The dynamics of photosystem I assembly in cyanobacteria have been addressed using in vivo pulse-chase labeling of Synechocystis sp. PCC 6803 proteins in combination with blue native polyacrylamide gel electrophoresis. The analyses indicate the existence of three different monomeric photosystem I complexes and also the high stability of photosystem I trimers. We show that in addition to a complete photosystem I monomer, containing all 11 subunits, we detected a PsaK-less monomer and a short-lived PsaL/PsaK-less complex. The latter two monomers were missing in the ycf37 mutant of Synechocystis sp. PCC 6803 that accumulates also less trimers. Pulse-chase experiments suggest that the three monomeric complexes have different functions in the biogenesis of the trimer. Based on these findings we propose a model where PsaK is incorporated in the latest step of photosystem I assembly. The PsaK-less photosystem I monomer may represent an intermediate complex that is important for the exchange of the two PsaK variants during high light acclimation. Implications of the presented data with respect to Ycf37 function are discussed. Photosystem I (PSI) 2The abbreviations used are: PSI, photosystem I; PSII, photosystem II; BN, blue native; Chl, chlorophyll a; PCC, Pasteur Culture Collection; [35S]Met, l-[35S]methionine; TES, 2-{[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]amino} ethanesulfonic acid; BisTris, 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol; Tricine, N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine. is a multisubunit pigment-protein complex of oxygenic photosynthesis located in the thylakoid membrane of chloroplasts and cyanobacteria. Its main function is the light-dependent electron transfer from plastocyanin or cytochrome c6 on the lumenal side to ferredoxin on the cytoplasmic or (in chloroplasts) on the stromal side of photosynthetic membranes. The crystal structures for cyanobacterial and plant PSI at 2.5 Å (1Jordan P. Fromme P. Witt H.T. Klukas O. Saenger W. Krauss N. Nature. 2001; 411: 909-917Crossref PubMed Scopus (2082) Google Scholar) and 4.4 Å (2Ben-Shem A. Frolow F. Nelson N. Nature. 2003; 426: 630-635Crossref PubMed Scopus (669) Google Scholar) resolution have improved the understanding of the function of this complex. The PSI subunits are designated PsaA-P, according to the corresponding psaA-P genes (3Jensen P.E. Haldrup A. Rosgaard L. Scheller H.V. Physiol. Plant. 2003; 119: 313-321Crossref Scopus (50) Google Scholar, 4Fromme P. Mathis P. Photosynth. Res. 2004; 80: 109-124Crossref PubMed Scopus (36) Google Scholar, 5Grotjohann I. Fromme P. Photosynth. Res. 2005; 85: 51-72Crossref PubMed Scopus (140) Google Scholar, 6Khrouchtchova A. Hansson M. Paakkarinen V. Vainonen J.P. Zhang S. Jensen P.E. Scheller H.V. Vener A.V. Aro E.M. Haldrup A. FEBS Lett. 2005; 579: 4808-4812Crossref PubMed Scopus (49) Google Scholar). Their primary structures are well conserved among eukaryotes and prokaryotes performing oxygenic photosynthesis with the exception of five subunits not present in cyanobacteria (PsaG, PsaH, PsaN, PsaO, and PsaP) and one subunit PsaM, which has not been detected in plant PSI preparations. A heterodimer of two integral membrane proteins, PsaA and PsaB, forms the core of the PSI reaction center, which binds the electron transfer components P700, A0,A1, and FX. The terminal electron acceptors of PSI, the [4Fe-4S] centers FA and FB are bound to PsaC. PsaD and PsaE form, together with PsaC, peripheral ridges on the cytoplasmic and the stromal side, respectively, that provides a ferredoxin-docking site (7Kruip J. Chitnis P.R. Lagoutte B. Rögner M. Boekema E.J. J. Biol. Chem. 1997; 272: 17061-17069Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 8Setif P. Fischer N. Lagoutte B. Bottin H. Rochaix J.D. Biochim. Biophys. Acta. 2002; 1555: 204-209Crossref PubMed Scopus (83) Google Scholar). PsaD is also required for the stable integration of PsaC, PsaL, and PsaE into the PSI complex (9Xu W. Tang H. Wang Y. Chitnis P.R. Biochim. Biophys. Acta. 2001; 1507: 32-40Crossref PubMed Scopus (26) Google Scholar). PsaE is a structural component involved in ferredoxin reduction and cyclic electron flow around PSI. PsaF is exposed to the lumenal side of the thylakoids. It functions as a plastocyanin, or in cyanobacteria cytochrome c6, docking site of PSI and is required for efficient electron transfer from these proteins to P700+ (10Farah J. Rappaport F. Choquet Y. Joliot P. Rochaix J.D. EMBO J. 1995; 14: 4976-4984Crossref PubMed Scopus (112) Google Scholar, 11Hippler M. Drepper F. Farah J. Rochaix J.D. Biochemistry. 1997; 36: 6343-6349Crossref PubMed Scopus (94) Google Scholar). PsaL is a subunit that forms most of the contacts between the monomers to form a trimer in cyanobacteria (12Karapetyan N.V. Holzwarth A.R. Rögner M. FEBS Lett. 1999; 460: 395-400Crossref PubMed Scopus (110) Google Scholar, 13Chitnis V.P. Chitnis P.R. FEBS Lett. 1993; 336: 330-334Crossref PubMed Scopus (194) Google Scholar). In contrast to cyanobacteria, trimeric PSI complexes have never been found in plants. Plant PsaL seems to be required for stable binding of PsaH and PsaO (3Jensen P.E. Haldrup A. Rosgaard L. Scheller H.V. Physiol. Plant. 2003; 119: 313-321Crossref Scopus (50) Google Scholar). PsaI plays a role in the structural organization of the PsaL subunit of PSI. PsaJ interacts with PsaF and seems to be important for stabilizing PsaF (3Jensen P.E. Haldrup A. Rosgaard L. Scheller H.V. Physiol. Plant. 2003; 119: 313-321Crossref Scopus (50) Google Scholar). Primary sequences of PsaK from higher plants and cyanobacteria do not show a high degree of similarity. PsaK in plants seems to be involved in the binding of light-harvesting complex I and in mediating state transitions (14Jensen P.E. Gilpin M. Knoetzel J. Scheller H.V. J. Biol. Chem. 2000; 275: 24701-24708Abstract Full Text Full Text PDF PubMed Scopus (122) Google Scholar, 15Scheller H.V. Jensen P.E. Haldrup A. Lunde C. Knoetzel J. Biochim. Biophys. Acta. 2001; 1507: 41-60Crossref PubMed Scopus (173) Google Scholar, 16Varotto C. Pesaresi P. Jahns P. Lessnick A. Tizzano M. Schiavon F. Salamini F. Leister D. Plant Physiol. 2002; 129: 616-624Crossref PubMed Scopus (60) Google Scholar), whereas in Synechocystis sp. PCC 6803 (hereafter Synechocystis 6803) it was shown that one of the two existing PsaK variants functions in high light acclimation (17Fujimori T. Hihara Y. Sonoike K. J. Biol. Chem. 2005; 280: 22191-22197Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar). The biosynthesis of the plant PSI complex depends on the coordinated expression of nuclear and plastid genes, the targeting of the subunits to their proper location within the chloroplast, the incorporation of the pigments and redox cofactors, and the correct assembly of the subunits to form a functionally active complex (18Rochaix J.D. FEBS Lett. 2002; 529: 34-38Crossref PubMed Scopus (74) Google Scholar, 19Rochaix J.D. Perron K. Dauvillee D. Laroche F. Takahashi Y. Goldschmidt-Clermont M. Biochem. Soc. Trans. 2004; 32: 567-570Crossref PubMed Scopus (16) Google Scholar). Additional gene products have been identified by mutational analyses that may act as regulatory or scaffold proteins in the assembly of PSI. The conserved chloroplast open reading frames Ycf3, Ycf4, and Ycf37 along with specific cyanobacterial proteins like BtpA and RubA appear to be involved in posttranslational steps of PSI assembly. Ycf3 was shown to interact with PsaA and PsaD (20Naver H. Boudreau E. Rochaix J.D. Plant Cell. 2001; 13: 2731-2745Crossref PubMed Scopus (108) Google Scholar), whereas Ycf4 is associated with a high molecular mass complex containing several PSI subunits (19Rochaix J.D. Perron K. Dauvillee D. Laroche F. Takahashi Y. Goldschmidt-Clermont M. Biochem. Soc. Trans. 2004; 32: 567-570Crossref PubMed Scopus (16) Google Scholar). Ycf3 as well as ycf4 mutants from Chlamydomonas lack PSI complexes (21Boudreau E. Takahashi Y. Lemieux C. Turmel M. Rochaix J.D. EMBO J. 1997; 16: 6095-6104Crossref PubMed Scopus (206) Google Scholar). A ycf4 mutant from Synechocystis 6803 still has functional PSI complexes but in lower amounts (22Wilde A. Härtel H. Hübschmann T. Hoffmann P. Shestakov S.V. Börner T. Plant Cell. 1995; 7: 649-658Crossref PubMed Scopus (53) Google Scholar). A cyanobacterial ycf37 mutant lacks a specific PsaK-less PSI monomeric complex and contains only 70% trimeric PSI complexes when compared with wild-type thylakoids (23Dühring U. Irrgang K.D. Lünser K. Kehr J. Wilde A. Biochim. Biophys. Acta. 2006; 1757: 3-11Crossref PubMed Scopus (44) Google Scholar). On the other hand mutation of the plant ycf37-homolog pyg7 leads to a complete loss of functional PSI in Arabidopsis thaliana (24Stöckel J. Bennewitz S. Hein P. Oelmüller R. Plant Physiol. 2006; 141: 870-878Crossref PubMed Scopus (54) Google Scholar). In cyanobacteria the cores of photosynthetic reaction centers are thought to be first assembled in the plasma membrane, before they are translocated to the thylakoid membrane. Several assembly factors and photosystem subunits were found to be present in the plasma membrane (25Zak E. Norling B. Maitra R. Huang F. Andersson B. Pakrasi H.B. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 13443-13448Crossref PubMed Scopus (149) Google Scholar, 26Klinkert B. Ossenbühl F. Sikorski M. Berry S. Eichacker L. Nickelsen J. J. Biol. Chem. 2004; 279: 44639-44644Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). In the case of PSI a heterodimer of PsaA and PsaB most likely forms the folding center of PSI and provides the scaffold for other protein subunits to assemble. However, the detailed mechanism of biogenesis of the photosynthetic apparatus and the molecular function of the various assembly factors remains poorly understood. The cyanobacterium Synechocystis 6803 is an excellent model system to study PSI assembly. Cyanobacterial mutants often still accumulate impaired PSI complexes, whereas similar mutations in eukaryotic algae and plants lead to a complete loss of the complex. Using blue native (BN)-PAGE in combination with in vivo pulse-chase labeling of proteins we addressed the dynamics of PSI protein synthesis and assembly in Synechocystis 6803 wild-type and in the ycf37 mutant. We show the existence of three different monomeric PSI complexes: a short-lived PsaL/PsaK-less assembly intermediate, a strongly labeled PsaK-less complex, and a third complete monomer that contains all 11 subunits. Lack of the short-lived assembly intermediate as well as the absence of the PsaK-less monomeric complex in the ycf37 mutant suggests a function of the Ycf37 protein in formation or stabilization of these assembly/disassembly intermediates. Bacterial Strains and Growth Conditions—Liquid cultures of Synechocystis 6803 wild-type (S. Shestakov, Moscow State University) and mutant strains were grown photoautotrophically at 30 °C in BG-11 medium (27Rippka R. Desrulles J. Waterbury J.B. Herdman M. Stanier R.Y. J. Gen. Microbiol. 1979; 11: 419-436Google Scholar) or photoheterotrophically (0.2% glucose) under continuous illumination with white light of 50 μmol photons m–2 s–1 and bubbled with air. The medium for the ycf37 (28Wilde A. Lünser K. Ossenbühl F. Nickelsen J. Börner T. Biochem. J. 2001; 357: 211-216Crossref PubMed Scopus (46) Google Scholar) and psaK2 (17Fujimori T. Hihara Y. Sonoike K. J. Biol. Chem. 2005; 280: 22191-22197Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar) mutant strains was supplemented with 7 μgml–1 chloramphenicol and 40 μgml–1 kanamycin, respectively. Pulse-Chase Labeling—Photoheterotrophically grown cells in the logarithmic growth phase were harvested by centrifugation, washed, and resuspended in BG11 medium (supplemented with 8 mm NaHCO3, 0.8 mm Na2CO3, 25 mm TES, pH 8.0) to a final volume of 1.5 ml with a chlorophyll (Chl) concentration of 200 μgml–1. For the ycf37 mutant the final Chl concentration was adjusted to 150 μgml–1 (75% of the wild-type Chl level) to accommodate for the lower Chl content of this strain (28Wilde A. Lünser K. Ossenbühl F. Nickelsen J. Börner T. Biochem. J. 2001; 357: 211-216Crossref PubMed Scopus (46) Google Scholar). The cell suspension was preincubated at 5 μmol photons m–2 s–1 for 30 min at 30 °C in a water bath shaker. After the addition of l-[35S]methionine ([35S]Met) to a final concentration of 500 μCi ml–1 (1000 Ci mmol–1; Amersham Biosciences), cells were incubated for another 20 min under the same conditions. The labeling reaction was stopped by removal of unincorporated [35S]Met by centrifugation. Furthermore, lincomycin (final concentration 0.5 mg ml–1) as well as a 1000-fold excess of non-radioactive l-Met was added to block translation and labeling. Aliquots were frozen in liquid nitrogen immediately. The remaining cells were illuminated either under high light (250 μmol m–2 s–1) or low light conditions (15 μmol m–2 s–1). Aliquots were taken after 1, 4, 8, and 22 h and rapidly frozen in liquid nitrogen. Thylakoids were prepared as described (23Dühring U. Irrgang K.D. Lünser K. Kehr J. Wilde A. Biochim. Biophys. Acta. 2006; 1757: 3-11Crossref PubMed Scopus (44) Google Scholar). BN-PAGE and Immunoblot Analysis—Thylakoid membranes (about 300 μg of protein) were sedimented by centrifugation (20 min, 15,000 × g, 4 °C) and resuspended in 50 μl of ACA buffer (50 mm BisTris/HCl, pH 7.0, 750 mm ϵ-amino-n-caproic acid, 0.5 mm EDTA). Membrane proteins were solubilized by the addition of 7 μl of freshly prepared 10% (w/v) n-dodecyl-β-d-maltoside solution and incubation for 5 min on ice. After centrifugation (15,000 × g, 30 min), the supernatants were supplemented with 10 μl of a Coomassie Blue solution (5% (w/v) Serva Brilliant Blue G-250, 750 mm ϵ-amino-n-caproic acid) and loaded onto a 7% BN-gel. One-dimensional BN-PAGE and BN/Tricine-urea-SDS-PAGE were carried out as described (29Jänsch L. Kruft V. Schmitz U.K. Braun H.P. Plant J. 1996; 9: 357-368Crossref PubMed Scopus (191) Google Scholar). If needed, bands of interest were excised from a one-dimensional BN-gel, placed on top of a resolving gel for SDS-PAGE, embedded in stacking gel, and subsequently separated by electrophoresis. Proteins were electrophoretically transferred onto nitrocellulose membranes and immunodecorated with antibodies against PsaC (AgriSera, Vännäs, Sweden), PsaD, PsaE, PsaF, and PsaL. Monitoring of PSI Assembly Intermediates—BN-PAGE is a powerful tool to investigate membrane-bound protein complexes (29Jänsch L. Kruft V. Schmitz U.K. Braun H.P. Plant J. 1996; 9: 357-368Crossref PubMed Scopus (191) Google Scholar, 30Schägger H. Cramer W.A. von Jagow G. Anal. Biochem. 1994; 217: 220-230Crossref PubMed Scopus (1043) Google Scholar). This technique separates protein complexes by their size without dissociating them into their individual polypeptides. The integration and assembly of different subunits of PSII could thus be followed by pulse-chase radiolabeling of cells in combination with two-dimensional BN/SDS-PAGE (31Komenda J. Reisinger V. Muller B.C. Dobakova M. Granvogl B. Eichacker L.A. J. Biol. Chem. 2004; 279: 48620-48629Abstract Full Text Full Text PDF PubMed Scopus (136) Google Scholar, 32Ossenbühl F. Gohre V. Meurer J. Krieger-Liszkay A. Rochaix J.D. Eichacker L.A. Plant Cell. 2004; 16: 1790-1800Crossref PubMed Scopus (101) Google Scholar, 33Ossenbühl F. Inaba-Sulpice M. Meurer J. Soll J. Eichacker L.A. Plant Cell. 2006; 18: 2236-2246Crossref PubMed Scopus (50) Google Scholar). Here we used a similar method to analyze PSI assembly. Since PSI subunits do not accumulate as much label as subunits of PSII reaction centers, Synechocystis 6803 cells were incubated with [35S]Met under low light conditions to achieve higher labeling level of PSI proteins. As these are limiting growth conditions for the cells we supplemented the media with glucose to enhance growth rate. Photoheterotrophic growth conditions led to a faster protein synthesis and therefore to an optimized incorporation of [35S]Met. After the radioactive pulse the turnover of trimeric PSI complexes was enhanced during the chase by illumination with high light to detect potential monomeric degradation products. To characterize the previously described PsaK-less monomeric PSI complex PSI* (23Dühring U. Irrgang K.D. Lünser K. Kehr J. Wilde A. Biochim. Biophys. Acta. 2006; 1757: 3-11Crossref PubMed Scopus (44) Google Scholar), wild-type and ycf37 mutant cells were analyzed. Isolated thylakoid membranes were subjected to BN-PAGE. Seven radiolabeled protein complexes with a different electrophoretic mobility could be detected (Fig. 1). Two complexes were missing in the ycf37 mutant, suggesting that one of these complexes represents PSI*. The PSI trimer was labeled less intensely in the ycf37 mutant, whereas the amount of radioactivity in PSII complexes was comparable in wild-type and mutant thylakoids. The amount of radiolabel in the PSI trimer increased slightly during the chase, while label in the PSII complexes decreased, mainly due to the enhanced damage of PSII complexes and inhibition of the PSII repair under high light conditions (34Nishiyama Y. Allakhverdiev S.I. Murata N. Biochim. Biophys. Acta. 2006; 1757: 742-749Crossref PubMed Scopus (567) Google Scholar). Interestingly, radiolabel accumulated significantly in the PSI monomer during the chase, whereas the amount of label in PSI* did not change in wild-type thylakoids (Fig. 1). These results indicate a higher stability of PSI trimers in comparison to PSII complexes. Furthermore, PSI* and PSI trimers seem to be stable products of PSI assembly. The regular PSI monomer may represent a degradation intermediate and/or a pool of monomeric PSI complexes for recycling of the trimeric complex under high light conditions. Besides the well documented PSI complexes (PSI trimer, PSI monomer, PsaK-less PSI monomer (PSI*)) an additional radiolabeled complex with a slightly higher electrophoretic mobility than PSI* could be detected in the wild type but not in the ycf37 mutant (Fig. 1). As we detect this complex which we call PSI** only in the autoradiograms but not after Coomassie staining of the BN-gels, and since label in this complex disappears during chase time, we conclude that it represents an unstable and/or short-lived assembly intermediate. The Subunit Composition of Different PSI Complexes—We found that the newly detected short-lived assembly intermediate PSI** is missing in the ycf37 mutant and that it migrates slightly faster than the two previously characterized PSI monomers (23Dühring U. Irrgang K.D. Lünser K. Kehr J. Wilde A. Biochim. Biophys. Acta. 2006; 1757: 3-11Crossref PubMed Scopus (44) Google Scholar). Thus, PSI** could represent a PSI subcomplex lacking specific subunits. To detect differences in the subunit composition between the individual monomeric PSI complexes, the lanes of the BN-PAGE were subjected to a second denaturating electrophoresis. The silver-stained two-dimensional BN/SDS-PAGE gel is shown as a reference indicating the presence and localization of the individual subunits of the major protein complexes (Fig. 2A). The corresponding autoradiogram in Fig. 2B depicts the [35S]Met-labeled newly synthesized translation products and the presence of assembly intermediates. All PSI subunits (except PsaE, which does not contain a Met) incorporated [35S]Met and were detected in the PSI trimer and the regular PSI monomer, as is evident from comparison of the silver stained gel with the autoradiogram. Due to their small size it was not possible to exactly assign the polypeptides PsaM, PsaI, and PsaJ (Fig. 2). The next PSI subcomplex with decreasing size (PSI*) was identified in silver-stained gels as well as in the autoradiograms. This is characterized by the lack of PsaK. In addition, PsaF was poorly labeled in all PSI monomer complexes during pulse time (although PsaF contains one more Met than e.g. PsaL). During chase monomeric as well as trimeric PSI complexes incorporated relatively more [35S]Met-labeled PsaF, suggesting that a larger pool of unassembled PsaF is present in the thylakoid membrane. This suggests that PsaF can be independently exchanged between existing PSI complexes and this pool. The smallest detectable assembly intermediate PSI** which is visible only in the pulse samples seems to lack PsaK as well as PsaL. To clarify the presence of other PSI subunits that are poorly detectable in the autoradiogram we performed immunoblot analysis (Fig. 3) using an enriched PSI** fraction. The region of a BN-gel where PSI** migrates corresponding to the autoradiogram was excised and subjected to a second dimension. Immunoblot analysis indicate the presence of PsaF, PsaE, PsaD, and PsaC, wheras PsaL is clearly missing in the PSI** complex.FIGURE 3Immunodetection of the PSI subunits PsaC, PsaD, PsaE, PsaF, and PsaL in monomeric PSI complexes. Protein complexes of solubilized wild-type thylakoid membranes were separated by BN-PAGE. The bands containing monomeric photosystem complexes were excised from the gel, embedded in the stacking gel, subjected to Tricine-urea-SDS-PAGE, and transferred to nitrocellulose membranes. Antibodies used for the immunoblots are indicated on the right. To achieve a 4-fold enrichment of the short lived complex, the PSI** band was excised from a lane loaded with 600 μg of protein (150 μg of protein for the other PSI monomer bands).View Large Image Figure ViewerDownload Hi-res image Download (PPT) Our results suggest a rapid assembly of PSI** into PSI trimers most likely via PSI* as an intermediate. We conclude that PsaL and then PsaK are the last subunits of PSI to be assembled. Synechocystis 6803 contains two different PsaK proteins, of which PsaK1 is dominant under low light conditions, while PsaK2 is more abundant under high light conditions (17Fujimori T. Hihara Y. Sonoike K. J. Biol. Chem. 2005; 280: 22191-22197Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar). Thus, the incorporation of PsaK1/PsaK2 as final PSI subunit might make sense as a possible acclimation step during recycling of PSI complexes under changing light conditions. To verify this hypothesis we analyzed the presence of PSI* in psaK2 mutants under various light intensities. High light conditions lead to an increase of PSI* as was shown previously (23Dühring U. Irrgang K.D. Lünser K. Kehr J. Wilde A. Biochim. Biophys. Acta. 2006; 1757: 3-11Crossref PubMed Scopus (44) Google Scholar). However, in the psaK2 mutant this accumulation is much more pronounced than in the wild type (Fig. 4), whereas low light did not induce an increase in the PSI* amount. Dynamics of PSI Complexes at Low and High Light Intensities—To follow the assembly/disassembly of PSI complexes under different light conditions we next performed in vivo labeling experiments with [35S]Met using different light conditions in the chase. Thylakoid membranes were isolated, solubilized, and thylakoid membrane complexes were analyzed by BN-PAGE (Fig. 5). Incubation at low light resulted in very stable photosystems. Accumulated radiolabel in assembled PSII monomers and dimers as well as trimeric PSI complexes remained constant over the whole chase period (Fig. 5A). In contrast, treating cells with high light led to degradation of photosynthetic complexes, in particular of PSII complexes due to its photoinactivation. However, no distinct accumulation of degradation products of the photosystems was detected during the chase at high light (Fig. 5B). In comparison to PSII and monomeric PSI complexes, PSI trimers showed the highest stability under high light conditions, with little decrease detectable up to 8 h of chase. Radiolabel in the regular PSI monomer was weak after the pulse in comparison to the other PSI complexes but increased continuously up to 4 h of chase. Possibly, this monomer represents an early intermediate of disintegrating trimers. In contrast to low light conditions after 4 h chase high light induced degradation of PSI and PSI* monomers (Fig. 5B). Thus, both complexes appear to be in a dynamic equilibrium with the trimers. Though structure and function of the PSI has been elucidated in detail (1Jordan P. Fromme P. Witt H.T. Klukas O. Saenger W. Krauss N. Nature. 2001; 411: 909-917Crossref PubMed Scopus (2082) Google Scholar, 5Grotjohann I. Fromme P. Photosynth. Res. 2005; 85: 51-72Crossref PubMed Scopus (140) Google Scholar, 35Chitnis P.R. Annu. Rev. Plant Physiol. Plant Mol. Biol. 2001; 52: 593-626Crossref PubMed Scopus (210) Google Scholar), its assembly, repair, and degradation are poorly understood. The functional significance of the recently discovered PSI monomer lacking the PsaK subunit (PSI*) (23Dühring U. Irrgang K.D. Lünser K. Kehr J. Wilde A. Biochim. Biophys. Acta. 2006; 1757: 3-11Crossref PubMed Scopus (44) Google Scholar) is also poorly established. In the current study the assembly and dynamics of PSI complexes have been analyzed in the wild type and the ycf37 mutant of the cyanobacterium Synechocystis 6803 using pulse-chase experiments. Labeling of cyanobacterial cells was followed by a BN-PAGE analysis to examine the dynamics of assembly and disassembly intermediates of PSI complexes. This approach allowed us to detect three specific monomeric PSI complexes with different labeling (i.e. turnover) kinetics. Surprisingly, we detected only late assembly intermediates. These results may either indicate that the early assembly steps are very fast or that the assembly intermediates may not be stable enough to detect with our method. As shown in our model (Fig. 6) the first assembly intermediate detected following the synthesis of individual subunits and formation of the PSI core is a PsaL/PsaK-less PSI monomer (PSI**). The next step is the integration of the PsaL subunit into the complex resulting in the recently discovered PSI* intermediate. Finally, addition of PsaK leads to a complete PSI monomer consisting of all 11 PSI subunits. Hippler et al. (36Hippler M. Rimbault B. Takahashi Y. Protist. 2002; 153: 197-220Crossref PubMed Scopus (20) Google Scholar) suggested that in Chlamydomonas PsaK is also the last subunit assembled into PSI complexes (36Hippler M. Rimbault B. Takahashi Y. Protist. 2002; 153: 197-220Crossref PubMed Scopus (20) Google Scholar). Pulse-chase experiments suggest a homeostasis of trimeric and the two "late" monomeric complexes PSI and PSI*. The question remains why PsaK and not PsaL is the last subunit that is incorporated into the complex? PsaL is essential (but not sufficient) for trimerization of PSI in cyanobacteria. Thus, association of PsaL with a PSI monomer as a final step in the assembly of trimers would agree with this role. However, in plants PSI is a monomeric complex, although it contains PsaL. We suggest that the association of PsaL with PSI subassemblies does not always directly lead to trimerization but may also depend on the attachment of PsaK. The cyanobacterium Synechocystis 6803 contains two psaK genes, psaK1 and psaK2. The psaK2 mRNA is the only transcript of a PSI gene that accumulates under high light conditions (37Hihara Y. Kamei A. Kanehisa M. Kaplan A. Ikeuchi M. Plant Cell. 2001; 13: 793-806Crossref PubMed Scopus (383) Google Scholar) suggesting that it is involved in responses to high light. PsaK has two transmembrane helices and is located peripherally, close to the interface between the monomers of a PSI trimer (1Jordan P. Fromme P. Witt H.T. Klukas O. Saenger W. Krauss N. Nature. 2001; 411: 909-917Crossref PubMed Scopus (2082) Google Scholar). It shares around 30% amino acid identity with PsaG, a plant PSI subunit that is lacking in cyanobacteria. Studies of psaG and psaK Arabidopsis mutants (14Jensen P.E. Gilpin M. Knoetzel J. Scheller H.V. J. Biol. Chem. 2000; 275: 24701-24708Abstract Full Text Full Text PDF PubMed Scopus (122) Google Scholar, 15Scheller H.V. Jensen P.E. Haldrup A. Lunde C. Knoetzel J. Biochim. Biophys. Acta. 2001; 1507: 41-60Crossref PubMed Scopus (173) Google Scholar, 16Varotto C. Pesaresi P. Jahns P. Lessnick A. Tizzano M. Schiavon F. Salamini F. Leister D. Plant Physiol. 2002; 129: 616-624Crossref PubMed Scopus (60) Google Scholar, 38Zygadlo A. Jensen P.E. Leister D. Scheller H.V. Biochim. Biophys. Acta. 2005; 1708: 154-163Crossref PubMed Scopus (19) Google Scholar) have suggested a stabilizing role of these subunits for the PSI core and the peripheral antenna. In plants PsaG and PsaK are incorporated only under low light conditions, suggesting a particular role of peripheral subunits in light acclimation (39Rokka A. Suorsa M. Saleem A. Battchikova N. Aro E.M. Biochem. J. 2005; 388: 159-168Crossref PubMed Scopus (149) Google Scholar). In cyanobacteria PsaK2 was shown to be involved in excitation energy transfer from phycobilisomes to PSI under high light conditions (17Fujimori T. Hihara Y. Sonoike K. J. Biol. Chem. 2005; 280: 22191-22197Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar). In addition, Fujimori et al. (17Fujimori T. Hihara Y. Sonoike K. J. Biol. Chem. 2005; 280: 22191-22197Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar) observed that the amount of the PsaK2 subunit increased in PSI complexes of high light acclimated cells. PsaG as well as PsaK are located on the outer edge of the plant PSI complex. Since the space that is occupied by PsaG in plant PSI is empty in cyanobacterial PSI as judged from the crystal structure, Fujimori et al. (17Fujimori T. Hihara Y. Sonoike K. J. Biol. Chem. 2005; 280: 22191-22197Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar) suggested that PsaK2 could occupy this site under high light conditions. However, in thylakoids of high light acclimated cells we could not detect a stable third monomeric PSI complex (besides the regular PSI monomer and PSI*) that would contain both PsaK subunits (data not shown). To support our idea that PSI* is involved in the PsaK exchange process we analyzed PSI* accumulation in a psaK2 mutant. As expected Fig. 4 shows accumulation of the PsaK-less monomer in psaK2 mutants under high light conditions whereas under low light conditions no differences to the wild type were detected. Thus, in our model PsaK2 is assembled into PSI* under high light conditions at the expense of PsaK1, which is present under normal growth light (Fig. 6). In psaK2 mutants this PSI intermediate accumulates because of the inability of the cells to integrate PsaK2 during the acclimation process. We therefore suggest that PSI* has a dual role as a stable assembly product as well as an intermediate for exchange of the different PsaK subunits. A similar mechanism is known for the PSII subunit D1: high light induces the exchange of two different D1 proteins in cyanobacteria (40Clarke A.K. Soitamo A. Gustafsson P. Öquist G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9973-9977Crossref PubMed Scopus (103) Google Scholar). However, the existence of multiple D1 copies is not typical for all cyanobacteria (e.g. several Prochlorococcus strains). The same holds true for the PsaK subunit. Fujimori et al. (17Fujimori T. Hihara Y. Sonoike K. J. Biol. Chem. 2005; 280: 22191-22197Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar) searched for the two types of PsaK proteins in other cyanobacteria. They found that cyanobacteria such as Synechococcus elongatus PCC 7942 and Trichodesmium erythraeum IMS 101 contain two types of the psaK gene, whereas the marine cyanobacterial strains Prochlorococcus and Synechococcus have only one psaK gene that forms a distinct clade in a phylogenetic tree. The Anabaena sp. PCC 7120 genome encodes even three psaK genes, one of the psaK1 type, whereas the other two are quite divergent from psaK2 and the marine type genes. These data imply that the assembly mechanism of different PsaK subunits into the PSI monomer is not specific to Synechocystis 6803 but is rather common in cyanobacterial strains that have to adapt to changing light conditions and that perform state transitions. However, the role of Ycf37 in the biogenesis of PSI is not specific to PsaK1/PsaK2 incorporation but has to be more vital, as this protein is found in all oxygenic photosynthetic organisms. Proteomic analysis of high light acclimation responses in Arabidopsis indicated a pronounced up-regulation of the Ycf37 homolog Pyg7 (41Giacomelli L. Rudella A. van Wijk K.J. Plant Physiol. 2006; 141: 685-701Crossref PubMed Scopus (124) Google Scholar). These data support our conclusion that this protein has a specific role in cyanobacterial high light acclimation (23Dühring U. Irrgang K.D. Lünser K. Kehr J. Wilde A. Biochim. Biophys. Acta. 2006; 1757: 3-11Crossref PubMed Scopus (44) Google Scholar). The inactivation of Pyg7 leads to a complete loss of all PSI complexes in Arabidopsis (24Stöckel J. Bennewitz S. Hein P. Oelmüller R. Plant Physiol. 2006; 141: 870-878Crossref PubMed Scopus (54) Google Scholar). In contrast, cyanobacterial ycf37 mutants contained fully assembled monomeric as well as trimeric PSI complexes, although in lower amounts. These results suggest that cyanobacteria are able to compensate more effectively for the lack of assembly factors, as was shown also for ycf4 (21Boudreau E. Takahashi Y. Lemieux C. Turmel M. Rochaix J.D. EMBO J. 1997; 16: 6095-6104Crossref PubMed Scopus (206) Google Scholar, 22Wilde A. Härtel H. Hübschmann T. Hoffmann P. Shestakov S.V. Börner T. Plant Cell. 1995; 7: 649-658Crossref PubMed Scopus (53) Google Scholar) and for several small PSI subunits (42Chitnis V.P. Xu Q. Yu L. Golbeck J.H. Nakamoto H. Xie D.L. Chitnis P.R. J. Biol. Chem. 1993; 268: 11678-11684Abstract Full Text PDF PubMed Google Scholar, 43Xu Q. Hoppe D. Chitnis V.P. Odom W.R. Guikema J.A. Chitnis P.R. J. Biol. Chem. 1995; 270: 16243-16250Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar, 44Naithani S. Hou J.M. Chitnis P.R. Photosynth. Res. 2000; 63: 225-236Crossref PubMed Scopus (28) Google Scholar). In addition, the Synechocystis Ycf37 protein as well as Pyg7 from Arabidopsis were found to bind to PSI complexes (23Dühring U. Irrgang K.D. Lünser K. Kehr J. Wilde A. Biochim. Biophys. Acta. 2006; 1757: 3-11Crossref PubMed Scopus (44) Google Scholar, 24Stöckel J. Bennewitz S. Hein P. Oelmüller R. Plant Physiol. 2006; 141: 870-878Crossref PubMed Scopus (54) Google Scholar). In cyanobacteria it is known that the early steps of photosystem assembly occur in the plasma membrane (25Zak E. Norling B. Maitra R. Huang F. Andersson B. Pakrasi H.B. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 13443-13448Crossref PubMed Scopus (149) Google Scholar). In contrast to other known PSI assembly factors like Ycf3 and Ycf4 that were found mainly in the plasma membrane, Ycf37 and also Pyg7 are located in the thylakoid membrane (24Stöckel J. Bennewitz S. Hein P. Oelmüller R. Plant Physiol. 2006; 141: 870-878Crossref PubMed Scopus (54) Google Scholar, 45Dühring U. Karradt A. Mikolajczyk S. Irrgang K.-D. Wilde A. van der Est A. Bruce D. Photosynthesis: Fundamental Aspects to Global Perspectives: Proceedings 13th International Congress on Photosynthesis. Alliance Communications Group Publishing, Montreal2005: 802-804Google Scholar). Therefore Ycf37 should rather act on later steps of PSI assembly. Most probably Ycf37 stabilizes late PSI assembly intermediates in the thylakoid membrane and protects them from premature degradation through proteolysis. We thank Gisa Baumert for technical assistance and Axel Brennicke, Mark Roberts, and Wolfgang Lockau for critical review of the manuscript. We thank Kintake Sonoike for providing the psaK2 mutant and Parag R. Chitnis for antibodies.
Referência(s)