Chimeric Photosynthetic Reaction Center Complex of Purple Bacteria Composed of the Core Subunits of Rubrivivax gelatinosus and the Cytochrome Subunit of Blastochloris viridis
2003; Elsevier BV; Volume: 278; Issue: 6 Linguagem: Inglês
10.1074/jbc.m209069200
ISSN1083-351X
AutoresHideaki Maki, Katsumi Matsuura, Keizo Shimada, Kenji V. P. Nagashima,
Tópico(s)Algal biology and biofuel production
ResumoA gene coding for the photosynthetic reaction center-bound cytochrome subunit, pufC, ofBlastochloris viridis, which belongs to the α-purple bacteria, was introduced into Rubrivivax gelatinosus, which belongs to the β-purple bacteria. The cytochrome subunit of B. viridis was synthesized in the R. gelatinosus cells, in which the native pufC gene was knocked out, and formed a chimeric reaction center (RC) complex together with other subunits ofR. gelatinosus. The transformant was able to grow photosynthetically. Rapid photo-oxidization of the hemes in the cytochrome subunit was observed in the membrane of the transformant. The soluble electron carrier, cytochrome c 2, isolated from B. viridis was a good electron donor to the chimeric RC. The redox midpoint potentials and the redox difference spectra of four hemes in the cytochrome subunit of the chimeric RC were almost identical with those in the B. viridis RC. The cytochrome subunit of B. viridis seems to retain its structure and function in the R. gelatinosus cell. The chimeric RC and its mutagenesis system should be useful for further studies about the cytochrome subunit of B. viridis. A gene coding for the photosynthetic reaction center-bound cytochrome subunit, pufC, ofBlastochloris viridis, which belongs to the α-purple bacteria, was introduced into Rubrivivax gelatinosus, which belongs to the β-purple bacteria. The cytochrome subunit of B. viridis was synthesized in the R. gelatinosus cells, in which the native pufC gene was knocked out, and formed a chimeric reaction center (RC) complex together with other subunits ofR. gelatinosus. The transformant was able to grow photosynthetically. Rapid photo-oxidization of the hemes in the cytochrome subunit was observed in the membrane of the transformant. The soluble electron carrier, cytochrome c 2, isolated from B. viridis was a good electron donor to the chimeric RC. The redox midpoint potentials and the redox difference spectra of four hemes in the cytochrome subunit of the chimeric RC were almost identical with those in the B. viridis RC. The cytochrome subunit of B. viridis seems to retain its structure and function in the R. gelatinosus cell. The chimeric RC and its mutagenesis system should be useful for further studies about the cytochrome subunit of B. viridis. reaction center high potential iron-sulfur protein 4-morpholinepropanesulfonic acid Photosynthetic energy conversion begins in pigment-protein complexes called photosynthetic reaction centers. The reaction center (RC)1 of purple bacteria is a membrane protein complex consisting of L, M, H, and cytochrome subunits. The L and M subunits, which are significantly homologous to each other, form a heterodimer in the membrane and bind bacteriochlorophylls, bacteriopheophytins, and quinines, which work in the transfer of electrons across the membrane (1Lancaster C.R.D. Ermler U. Michel H. Blankenship R.E. Madigan M.T. Bauer C.E. Anoxygenic Photosynthetic Bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands1995: 503-526Google Scholar). The cytochrome subunit, which contains four c-type hemes, is bound to the LM core at the periplasmic side and works as the electron donor to the photooxidized special pair of bacteriochlorophylls in the RC (1Lancaster C.R.D. Ermler U. Michel H. Blankenship R.E. Madigan M.T. Bauer C.E. Anoxygenic Photosynthetic Bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands1995: 503-526Google Scholar). One of two quinone molecules bound to the RC diffuses out of the RC after receiving two electrons and two protons. The cytochromebc 1 complex and water-soluble electron carrier proteins, such as cytochrome c 2, mediate the electron transfer from the doubly reduced quinone (quinol) molecule to the oxidized c-type hemes in the cytochrome subunit (2Meyer T.E. Donohue T.J. Blankenship R.E. Madigan M.T. Bauer C.E. Anoxygenic Photosynthetic Bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands1995: 725-745Google Scholar). Some species, such as Rhodobacter sphaeroides, Rhodospirillum rubrum, and Rhodopseudomonas palustris, have another type of RC complex in which the cytochrome subunit is absent. The L and M subunits of RC are encoded by the pufL andpufM genes, respectively, which form an operon calledpuf together with the pufB and pufAgenes that encode the β and α subunits of the light-harvesting 1 complex. In species containing the RC-bound cytochrome subunit, thepufC gene coding for this subunit is located immediately downstream of the pufM gene. Amino acid sequences derived from the nucleotide sequences of the pufC genes of purple bacteria contain four CXXCH sequence motifs to bind c-type hemes (3Nagashima K.V.P. Sakuragi Y. Shimada K. Matsuura K. Photosynth. Res. 1998; 55: 349-355Crossref Google Scholar), with an exception in the Rhodovulum species (4Masuda S. Yoshida M. Nagashima K.V.P. Shimada K. Matsuura K. J. Biol. Chem. 1999; 274: 10795-10801Abstract Full Text Full Text PDF PubMed Scopus (33) Google Scholar). The hemes have been numbered as heme-1 to heme-4 according to the order of the appearance of the respective binding motifs in the amino acid sequence (5Weyer K.A. Lottspeich F. Gruenberg H. Lang F. Oesterhelt D. Michel H. EMBO J. 1987; 6: 2197-2202Crossref PubMed Google Scholar). The RC complex purified from a purple nonsulfur bacterium belonging to the α-subclass, Blastochloris viridis (formally calledRhodopseudomonas viridis) (6Woese C.R. Stackebrandt E. Weisburg W.G. Paster B.J. Madigan M.T. Fowler V.J. Hahn C.M. Blanz P. Gupta R. Nealson K.H. Fox G.E. System. Appl. Microbiol. 1984; 5: 315-326Crossref PubMed Scopus (232) Google Scholar, 7Hiraishi A. Int. J. Syst. Bacteriol. 1997; 47: 217-219Crossref PubMed Scopus (60) Google Scholar), has been crystallized (8Michel H. J. Mol. Biol. 1982; 158: 567-572Crossref PubMed Scopus (292) Google Scholar, 9Deisenhofer J. Epp O. Miki K. Huber R. Michel H. Nature. 1985; 318: 618-624Crossref PubMed Scopus (2576) Google Scholar, 10Deisenhofer J. Epp O. Sinning I. Michel H. J. Mol. Biol. 1995; 246: 429-457Crossref PubMed Scopus (563) Google Scholar). Its crystal structure showed that the four c-type hemes are arranged in a roughly linear manner, almost perpendicularly to the membrane. The spectrophotometric and thermodynamic properties of the four hemes and their arrangement in the structure have been extensively studied (11Dracheva S.M. Drachev L.A. Konstantinov A.A. Semenov A.Y. Skulachev V.P. Arutjunjan A.M. Shuvalov V.A. Zeberezhnaya S.M. Eur. J. Biochem. 1988; 171: 253-264Crossref PubMed Scopus (148) Google Scholar, 12Nitschke W. Rutherford A.W. Biochemistry. 1989; 28: 3161-3168Crossref Scopus (63) Google Scholar, 13Fritzsch G. Buchanan S. Michel H. Biochim. Biophys. Acta. 1989; 977: 157-162Crossref Scopus (49) Google Scholar, 14Shinkarev V.P. Drachev A.L. Dracheva S.M. FEBS Lett. 1990; 261: 11-13Crossref Scopus (18) Google Scholar, 15Alegria G. Dutton P.L. Biochim. Biophys. Acta. 1991; 1057: 258-272Crossref PubMed Scopus (46) Google Scholar, 16Fritz F. Moss D.A. Mäntele W. FEBS Lett. 1992; 297: 167-170Crossref PubMed Scopus (14) Google Scholar). The four hemes have high-low-high-low midpoint potentials from the special pair to the periplasmic space. They are distinguishable in terms of the peak positions in their α-absorption bands. A gene manipulation system for B. viridis was established previously (17Laußermair E. Oesterhelt D. EMBO J. 1992; 11: 777-783Crossref PubMed Scopus (32) Google Scholar), and some studies had been done using site-directed mutagenesis manipulated into the LM core complex and into the cytochrome subunit of B. viridis (18Chen I.P. Mathis P. Koepke J. Michel H. Biochemistry. 2000; 39: 3592-3602Crossref PubMed Scopus (38) Google Scholar). B. viridiscannot be inoculated aerobically and grows with a doubling time of 24 h under microaerobic respiratory conditions (17Laußermair E. Oesterhelt D. EMBO J. 1992; 11: 777-783Crossref PubMed Scopus (32) Google Scholar). The very low growth rate makes it difficult to obtain a sufficient number of the mutant strains of B. viridis. The expression of the B. viridis cytochrome subunit gene in Escherichia colicells was also challenged. However, it resulted in the accumulation of the inclusion body without heme incorporation (19Grisshammer R. Oeckl C. Michel H. Biochim. Biophys. Acta. 1991; 1088: 183-190Crossref PubMed Scopus (21) Google Scholar). The cytochrome subunit of Rubrivivax gelatinosus, which belongs to the β-subclass of purple bacteria, has also been studied and shown to be similar to that of B. viridis in terms of the redox potentials of hemes (20Nitschke W. Agalidis I. Rutherford A.W. Biochim. Biophys. Acta. 1992; 1100: 49-57Crossref Scopus (26) Google Scholar). R. gelatinosus grows well under respiratory conditions (about 2 h of the doubling time) as well as under photosynthetic conditions (about 2.5 h of the doubling time); therefore, it is easy to introduce mutations into the photosynthetic apparatus (21Nagashima K.V.P. Shimada K. Matsuura K. FEBS Lett. 1996; 385: 209-213Crossref PubMed Scopus (26) Google Scholar). The photosynthetic growth rate of theR. gelatinosus mutant lacking the cytochrome subunit was about a half of that of the wild type, showing that the cytochrome subunit is not essential but is advantageous for photosynthesis (21Nagashima K.V.P. Shimada K. Matsuura K. FEBS Lett. 1996; 385: 209-213Crossref PubMed Scopus (26) Google Scholar). Based on the mutational replacements of charged amino acid residues distributed on the surface of the subunit, Osyczka et al.(22Osyczka A. Nagashima K.V.P. Sogabe S. Miki K. Yoshida M. Shimada K. Matsuura K. Biochemistry. 1998; 37: 11732-11744Crossref PubMed Scopus (30) Google Scholar, 23Osyczka A. Nagashima K.V.P. Shimada K. Matsuura K. Biochemistry. 1999; 38: 2861-2865Crossref PubMed Scopus (24) Google Scholar, 24Osyczka A. Nagashima K.V.P. Sogabe S. Miki K. Shimada K. Matsuura K. Biochemistry. 1999; 38: 15779-15790Crossref PubMed Scopus (24) Google Scholar) have shown that the low potential heme-1 located at the most distant position from the special pair is a direct electron acceptor from the soluble electron carrier, which suggests that all four hemes are involved in the electron transfer to the special pair. Several water-soluble electron carrier proteins are present in the periplasmic space of R. gelatinosus. Three of these proteins (high potential iron-sulfur protein (HiPIP), high potential cytochromec 8, and low potential cytochromec 8), have been shown to work as the electron donors to the RC-bound cytochrome (25Schoepp B. Parot P. Menin L. Gaillard J. Richaud P. Verméglio A. Biochemistry. 1995; 34: 11736-11742Crossref PubMed Scopus (71) Google Scholar, 26Osyczka A. Yoshida M. Nagashima K.V.P. Shimada K. Matsuura K. Biochim. Biophys. Acta. 1997; 1321: 93-99Crossref Scopus (19) Google Scholar, 27Menin L. Yoshida M. Jaquinod M. Nagashima K.V.P. Matsuura K. Parot P. Verméglio A. Biochemistry. 1999; 38: 15238-15244Crossref PubMed Scopus (21) Google Scholar). The physiological electron donor to the B. viridis RC is the cytochromec 2. The ionic strength dependence studies on the reaction of the cytochrome c 2 with theB. viridis RC have suggested that the heme-1 of the cytochrome subunit is the primary electron acceptor from the cytochrome c 2 (28Knaff D.B. Wille A. Long J.E. Kriauciunas A. Durham B. Millett F. Biochemistry. 1991; 30: 1303-1310Crossref PubMed Scopus (55) Google Scholar). However, direct evidence for the precise interaction mechanisms between the soluble electron donor and heme-1 in B. viridis is not yet available. In this study, the gene for the cytochrome subunit of R. gelatinosus was replaced by that of B. viridisusing a common restriction site at a conserved DNA region in both species. The mutagenized cells of R. gelatinosus produced a chimeric RC complex containing the cytochrome subunit derived fromB. viridis and the other subunits from the host. The chimeric RC was isolated and characterized. R. gelatinosus strain IL144RL2, which is a spontaneous mutant of the wild-type strain IL144 and shows greatly depressed production of the light-harvesting 2 complex, was used as a host strain for gene manipulations in this study. The R. gelatinosus strain ΔC constructed in a previous study was also used in this study, in which thepufC gene was knocked out by insertion of a kanamycin resistance gene (22Osyczka A. Nagashima K.V.P. Sogabe S. Miki K. Yoshida M. Shimada K. Matsuura K. Biochemistry. 1998; 37: 11732-11744Crossref PubMed Scopus (30) Google Scholar). R. gelatinosus strains were grown under aerobic-dark or anaerobic-light conditions at 30 °C with a PYS medium (21Nagashima K.V.P. Shimada K. Matsuura K. FEBS Lett. 1996; 385: 209-213Crossref PubMed Scopus (26) Google Scholar). B. viridis cells were grown under anaerobic-light conditions at 30 °C in a malate-basal salt medium containing 0.5% sodium malate, 0.1% yeast extract (Difco), 0.1% ammonium sulfate, 1% basal salt solution (29Nagashima K.V.P. Hiraishi A. Shimada K. Matsuura K. J. Mol. Evol. 1997; 45: 131-136Crossref PubMed Scopus (129) Google Scholar), 0.1% vitamin solution (29Nagashima K.V.P. Hiraishi A. Shimada K. Matsuura K. J. Mol. Evol. 1997; 45: 131-136Crossref PubMed Scopus (129) Google Scholar), and 20 mm sodium phosphate (pH 7.0). E. coli strain JM109 was used as a host for plasmids and grown aerobically with a Luria-Bertani medium or an SOB medium at 37 °C (30Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). When needed, ampicillin, kanamycin, or chloramphenicol was added to the E. coli and R. gelatinosus cultures at a final concentration of 50 μg/ml. The plasmid pGELPUF containing the 4.6-kb whole puf operon of R. gelatinosus was created by connecting two DNA fragments, pGH37 and pGI7, previously cloned from the genomic DNA library of R. gelatinosus strain IL144 (31Nagashima K.V.P. Matsuura K. Ohyama S. Shimada K. J. Biol. Chem. 1994; 269: 2477-2484Abstract Full Text PDF PubMed Google Scholar). The 3.7-kb region flanked byNotI restriction sites within the puf operon ofR. gelatinosus was replaced by an ampicillin resistance gene. This plasmid was named pΔPUF2 (Fig. 1 A) and used for construction of the R. gelatinosus mutant lacking thepuf operon. A 2.5-kb DNA fragment flanked byBamHI restriction sites and containing the 3′-region ofpufM and the entire pufC of B. viridiswas cloned from the genomic library of B. viridis DNA using a pUC119 plasmid as a cloning vector. This plasmid was named p9VP3 (Fig. 1 B). The nucleotide sequence of pufC in the p9VP3 plasmid was confirmed to be identical to that previously reported for the B. viridis pufC (5Weyer K.A. Lottspeich F. Gruenberg H. Lang F. Oesterhelt D. Michel H. EMBO J. 1987; 6: 2197-2202Crossref PubMed Google Scholar). Both the genes for the M subunits of R. gelatinosus and B. viridis have a common SalI restriction site at the conserved 3′-regions. The region downstream of this SalI site in the insert DNA of the plasmid pGELPUF was removed and replaced by the DNA fragment containing the region downstream of the corresponding SalI site of the p9VP3 plasmid, in which the entire pufC ofB. viridis was contained (Fig. 1 B). The DNA fragment containing this “chimeric” puf operon was cloned into a pHSG298 plasmid. The p2GPVC-KF plasmid was thus obtained. A frameshift did not occur in this “chimeric” pufoperon. The presumed pufM product derived from this plasmid was expected to have the N-terminal 292 amino acids of the R. gelatinosus M subunit and the C-terminal 34 amino acids from theB. viridis M subunit. DNA manipulation and hybridization were carried out according to a manual on molecular cloning (30Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar) or according to instructions supplied by the enzyme manufacturers. Plasmid DNAs were introduced into R. gelatinosus cells by means of electroporation, according to the methods previously described (21Nagashima K.V.P. Shimada K. Matsuura K. FEBS Lett. 1996; 385: 209-213Crossref PubMed Scopus (26) Google Scholar). Cells in the late logarithmic growth phase were harvested by centrifugation at 7,200 × g for 20 min, washed, and suspended in 10 mm MOPS-NaOH (pH 7.0). The cells were disrupted by passage through a French pressure cell at 1100–1200 kg/cm2 in the presence of a few grains of DNase I. The disrupted cell suspension was centrifuged at 28,000 × g for 15 min to remove the debris and ultracentrifuged at 340,000 × g for 20 min. The sediment was used as the membrane preparation. The cytochromec 2 of B. viridis and HiPIP ofR. gelatinosus were isolated from the remaining supernatant according to the methods previously reported (26Osyczka A. Yoshida M. Nagashima K.V.P. Shimada K. Matsuura K. Biochim. Biophys. Acta. 1997; 1321: 93-99Crossref Scopus (19) Google Scholar). The protein content in the membrane was determined with a Protein Assay Kit (Bio-Rad), using bovine serum albumin (fraction V) as a standard protein. Reduced-minus-oxidized difference spectra were measured with a UV-3000 spectrophotometer (Shimadzu). Two millimolar (final concentration) of potassium ferricyanide or a few grains of sodium dithionite were added for the oxidation or the reduction of the membrane preparations. SDS-PAGE was carried out according to Laemmli (32Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207537) Google Scholar). The proteins with a concentration of 5 mg/ml were denatured in 1% SDS by boiling for 1 min. Ten micrograms of protein was applied to each lane of the gel. After electrophoresis, the gel was treated with Coomassie Brilliant Blue using a Quick-CBB Kit (Wako, Japan) for staining the proteins or treated with TMBZ for detection of the hemes according to the method of Thomas et al. (33Thomas P.E. Ryan D. Levin W. Anal. Biochem. 1976; 75: 168-176Crossref PubMed Scopus (897) Google Scholar). The membranes prepared were suspended in a buffer containing 10 mmMOPS-NaOH (pH 7.0) to give an absorbance of 40 at 875 nm. The suspension was mixed with an equal volume of precooled 1.6% (v/v) β-d-octylthioglucoside and incubated on ice for 40 min with stirring. Subsequently, the suspension was applied on the top of a sucrose density gradient (from top to bottom, 10–40%, w/v) containing 10 mm MOPS-NaOH (pH 7.0) and 0.05% Triton X-100 and ultracentrifuged at 146,000 × g for 20 h. A brownish red band that formed at a sucrose density of about 25% was recovered as an RC fraction and concentrated with Centriflo CF25 membrane cones (Amicon). The RC fraction was diluted with a buffer containing 100 mm KCl, 20 mm MOPS-KOH (pH 7.0), and 0.05% Triton X-100. Xenon flash-induced absorption changes were measured with a single beam spectrophotometer designed and assembled in our laboratory (34Matsuura K. Shimada K. Biochim. Biophys. Acta. 1986; 852: 9-18Crossref Scopus (38) Google Scholar). The membranes were suspended in a 20 mm MOPS-KOH buffer (pH 7.0) containing 100 mmKCl, 1 mm sodium ascorbate, and 10 μm DAD. The concentrations of the samples were adjusted to give an absorbance of 1 at 860 nm for the R. gelatinosus strain ΔC, at 875 nm for other R. gelatinosus strains, and at 1012 nm forB. viridis. One micromolar valinomycin was added to the membrane preparation of the R. gelatinosus strain ΔC to reduce the effect of the band shift of carotenoids responding to the membrane potential. The kinetics of an electron transfer from the soluble electron donor to the RC was measured in a buffer containing 5 mm MOPS-NaOH (pH 7.0). The R. gelatinosus HiPIP and the B. viridis cytochrome c 2 were added at the final concentrations of 1 μm. One millimolar sodium ascorbate, 10 μm DAD, and 0.03% Triton X-100 were also added. Redox titration of hemes in the RC-bound cytochrome c was carried out as described previously (35Dutton P.L. Biochim. Biophys. Acta. 1971; 226: 63-80Crossref PubMed Scopus (347) Google Scholar) using a double beam spectrophotometer UV-3000 (Shimadzu). The redox potential was monitored with an Ag/AgCl electrode (TOA) connected to an HM-40S voltmeter (TOA). The concentration of the sample was adjusted to give an absorbance of 2 at 875 nm for R. gelatinosus and at 1012 nm for B. viridis. The buffer for the measurements contained 100 mm KCl, 20 mm MOPS-KOH (pH 7.0), and 0.05% Triton X-100. The redox mediators used are as follows: 100 μm potassium ferrocyanide, 10 μm DAD, 100 μm Fe-EDTA, 10 μm phenazine methosulfate, 10 μm vitamin K3, 10 μm2-hydroxy-1,4-naphthoquinone, and 3 μm pyocyanine. The oxygen was eliminated by a continuous stream of nitrogen gas. At the beginning of the measurements, the suspension was reduced by the addition of a solution of sodium ascorbate or sodium dithionite. Subsequently, the redox potential was raised stepwise by additions of a solution of potassium ferricyanide. The plasmid pΔPUF2, which contains a kanamycin resistance gene and the R. gelatinosus puf operon disrupted by the insertion of an ampicillin resistance gene (Fig. 1 A), was introduced into the cells of the R. gelatinosus strain IL144RL2. A strain showing resistance to ampicillin but sensitivity to kanamycin was picked up after successive cultivation under the pressure of ampicillin and named DP2. An absorption spectrum of the membrane prepared from the strain DP2 cells showed no accumulation of the RC-light-harvesting 1 complex (Fig. 2). Southern hybridization, PCR, and DNA sequencing experiments for the genomic DNA isolated from the strain DP2 showed that the 3.7-kb NotI-NotI DNA region containing pufBALMC was replaced by the ampicillin resistance gene (data not shown), probably due to a double cross-over recombination between the genomic DNA and the pΔPUF2 plasmid. The strain DP2 could not grow photosynthetically. The plasmid p2GPVC-KF, which contains the majority of thepuf genes of R. gelatinosus and the entirepufC of B. viridis (Fig. 1), was introduced into the R. gelatinosus strain DP2 cells. A new strain showing resistance to kanamycin was obtained and named VC-F. The VC-F cells were able to grow under anaerobic light conditions, although the growth rate was about 3 times slower than that of the strain IL144RL2. An absorption spectrum of the strain VC-F membrane was nearly identical with that of the strain IL144RL2 membrane (Fig. 2), suggesting that the strain VC-F recovered the ability to synthesize the RC-light-harvesting 1 complex without significant alterations in the environments of the photosynthetic pigments. The presence of puf genes in the strain VC-F cells was confirmed by Southern hybridization experiments. The p2GPVC-KF plasmid was detected in the cell lysate of the strain VC-F and seemed to be reproducible in the VC-F cells. The synthesis of the RC-bound cytochrome subunit derived from theB. viridis pufC gene in the cells of the R. gelatinosus strain VC-F was examined by SDS-PAGE of the membranes, as shown in Fig. 3. The band patterns stained with Coomassie Brilliant Blue were almost identical among the membrane samples of the three strains of R. gelatinosus. Minor differences between the patterns lower than 14 kDa for the strain ΔC and for two other strains, IL144RL2 and VC-F, may reflect the difference in the contents of the light-harvesting 2 complexes. In the gel stained for hemes, a peptide with a molecular mass of 43 kDa was detected in the membrane of the strain IL144RL2, which was assigned to the cytochrome subunit (36Fukushima A. Matsuura K. Shimada K. Satoh T. Biochim. Biophys. Acta. 1988; 933: 399-405Crossref Scopus (49) Google Scholar). No heme-stained peptides were observed in the membrane of the strain ΔC, which lacks the cytochrome subunit (22Osyczka A. Nagashima K.V.P. Sogabe S. Miki K. Yoshida M. Shimada K. Matsuura K. Biochemistry. 1998; 37: 11732-11744Crossref PubMed Scopus (30) Google Scholar). In the membrane of the strain VC-F, a heme-stained peptide was detected at the 39-kDa molecular mass, which coincided with that of the RC-bound cytochrome subunit of B. viridis. This result suggested that the RC-bound cytochrome subunit derived from B. viridis was expressed in the R. gelatinosus strain VC-F. Fig. 4 shows the reduced-minus-oxidized difference spectrum in the α-absorption band region of the cytochromes c in the membrane of the strain VC-F, together with those of the strains IL144RL2 and ΔC ofR. gelatinosus and B. viridis. The difference spectrum of the strain IL144RL2 membrane, which has two low potential hemes that peaked at 551 nm and two high potential hemes that peaked at 555 nm in the cytochrome subunit, showed an absorption maximum at 552 nm, which was consistent with the results obtained from the wild type of R. gelatinosus shown in a previous study (36Fukushima A. Matsuura K. Shimada K. Satoh T. Biochim. Biophys. Acta. 1988; 933: 399-405Crossref Scopus (49) Google Scholar). In the strain ΔC, the absorption bands derived from cytochromes were very small, as expected (21Nagashima K.V.P. Shimada K. Matsuura K. FEBS Lett. 1996; 385: 209-213Crossref PubMed Scopus (26) Google Scholar). The broad and small bands detected in the strain ΔC membrane were probably due to the cytochromebc 1 complex. The membrane of the strain VC-F, on the other hand, showed a different spectrum with an absorption maximum at 553 nm with a shoulder around 559 nm, which was similar to that obtained for the membrane of B. viridis. Hemes contained in the cytochrome subunit of the R. gelatinosus strain VC-F were further characterized by redox titration (Fig. 5). Four redox components were detected when the peak heights of the α-absorption bands of the cytochromes c in the membrane preparation were plotted against the ambient potentials. When the data were fitted with Nernst equations, the redox midpoint potentials (E m values) of these components were estimated to be −60, +32, +320, and +419 mV in the membrane and −60, +32, +310, and +398 mV in the solubilized RC complex (Fig. 5). These values were consistent with those reported for the four hemes in the B. viridis cytochrome subunit (TableI). Fig. 6shows the reduced-minus-oxidized difference spectra of the four components in the strain VC-F membrane. Four c-type hemes with different absorption maxima in the α-band regions were resolved. The heme with the highest E m value showed a narrow α-band peaked at 559 nm with a shoulder at near 552 nm. The heme with the second highest E m had a somewhat broad α-band peaked at 556 nm. The two low potential hemes showed narrow α-bands with peaks at 552 and 553 nm. These spectral and redox characteristics in the strain VC-F were almost identical with those reported forB. viridis (11Dracheva S.M. Drachev L.A. Konstantinov A.A. Semenov A.Y. Skulachev V.P. Arutjunjan A.M. Shuvalov V.A. Zeberezhnaya S.M. Eur. J. Biochem. 1988; 171: 253-264Crossref PubMed Scopus (148) Google Scholar). This indicated that the environments surrounding the hemes in the cytochrome subunit in the strain VC-F were almost the same as those in the original RC complex of B. viridis.Table IRedox and spectral properties of hemes in cytochrome subunits of R. gelatinosus and B. viridisHeme1-aNumbering of hemes was based on the appearance in the amino acid sequence (5).Reference/source1234Cytochrome subunit of R. gelatinosus peak position in α-band (nm)1-bWavelength of absorption maximum in the redox difference spectrum measured in this study.551555555551 R. gelatinosus Strain IL144RL2+64+281+348+64This work Fukushima et al.+90+330+330+90Ref. 36Fukushima A. Matsuura K. Shimada K. Satoh T. Biochim. Biophys. Acta. 1988; 933: 399-405Crossref Scopus (49) Google Scholar Nitschke et al.+70+300+320+130Ref. 20Nitschke W. Agalidis I. Rutherford A.W. Biochim. Biophys. Acta. 1992; 1100: 49-57Crossref Scopus (26) Google ScholarCytochrome subunit of B. viridis peak position in α-band (nm)1-bWavelength of absorption maximum in the redox difference spectrum measured in this study.553556559552 R. gelatinosus Strain VC-F−60+310+398+32This work B. viridis Wild type−73+310+384+35This work Dracheva et al.−60+310+380+20Ref. 11Dracheva S.M. Drachev L.A. Konstantinov A.A. Semenov A.Y. Skulachev V.P. Arutjunjan A.M. Shuvalov V.A. Zeberezhnaya S.M. Eur. J. Biochem. 1988; 171: 253-264Crossref PubMed Scopus (148) Google Scholar Nitschke and Rutherford−80+320+400+20Ref.12Nitschke W. Rutherford A.W. Biochemistry. 1989; 28: 3161-3168Crossref Scopus (63) Google Scholar Frizsch et al.−60+300+370+10Ref. 13Fritzsch G. Buchanan S. Michel H. Biochim. Biophys. Acta. 1989; 977: 157-162Crossref Scopus (49) Google Scholar Shinkarev et al.−50+312+360+20Ref. 14Shinkarev V.P. Drachev A.L. Dracheva S.M. FEBS Lett. 1990; 261: 11-13Crossref Scopus (18) Google Scholar Alegria and Dutton−51+295+370+50Ref.15Alegria G. Dutton P.L. Biochim. Biophys. Acta. 1991; 1057: 258-272Crossref PubMed Scopus (46) Google Scholar Friz et al.−49+321+383+33Ref.16Fritz F. Moss D.A. Mäntele W. FEBS Lett. 1992; 297: 167-170Crossref PubMed Scopus (14) Google Scholar1-a Numbering of hemes was based on the appearance in the amino acid sequence (5Weyer K.A. Lottspeich F. Gruenberg H. Lang F. Oesterhelt D. Michel H. EMBO J. 1987; 6: 2197-2202Crossref PubMed Google Scholar).1-b Wavelength of absorption maximum in the redox difference spectrum measured in this study. Open table in a new tab Figure 6Reduced-minus-oxidized difference spectra of cytochromes c in the membrane of the R. gelatinosus strain VC-F poised at various redox potentials. The concentration of the sample was adjusted to give an absorbance of 2 at 875 nm.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The redox and spectroscopic properties of the four hemes in the membrane preparations of the R. gelatinosus strain IL144RL2 are summarized in Table I. The values of the redox midpoint potentials in R. gelatinosus IL144RL2 were +64 mV for the two low potential hemes peaking at 551 nm and +281 and +348 mV for the two high potential hemes peaking at 555 nm. These E m values were consistent with those for the R. gelatinosusstrain IL144 (36Fukushima A. Matsuura K. Shimada K. Satoh T. Biochim. Biophys. Acta. 1988; 933: 399-405Crossref Scopus (49) Google Scholar) and for the R. gelatinosus strain 52 (20). Thus, the redox midpoint potentials of the four hemes determined in the strain VC-F and their spectra in the α-band region were clearly different from those in R. gelatinosus IL144RL2. Fig. 7 shows flash-induced absorption changes in the α-band region of the cytochromes c in the membranes prepared from the cells of R. gelatinosus strains and B. viridis. For these measurements, the two high potential hemes in the cytochrome subunit were reduced prior to the flash activation by the addition of 1 mm sodium ascorbate. In the membrane of the strain VC-F, rapid oxidation of the cytochrome was observed, as in the membranes of R. gelatinosus IL144RL2 and B. viridis. The flash-induced spectra derived from this rapid oxidation showed a peak at 556 nm in the membranes of the strain VC-F and B. viridis, indicating that the second high potential heme, c556 (heme-2), was oxidized by the flash. In the flash-induced absorption change of the cytochromes in the membrane of the strains IL144RL2, the major oxidized component showed a peak at 553 nm. This suggested that the low potential hemes peaked at 551 nm in the IL144RL2 cytochrome subunit were partially reduced prior to the flash activation possibly due to their relatively high redox potentials. No remarkable peaks were observed in the flash-induced spectrum in the membrane preparation of the strain ΔC due to the lack of the cytochrome subunit. Fig. 8 shows the kinetics of the electron transfer from the soluble electron carrier proteins to the cytochrome subunit in the prepared membrane. When the R. gelatinosusHiPIP was added, re-reduction of the cytochrome was observed with a second order rate constant of 1.82 × 10−7 in the IL144RL2 membrane but not in the membranes of the strain VC-F andB. viridis. This suggested that HiPIP, which is the physiologically main electron donor to the oxidized cytochrome subunit in the R. gelatinosus wild-type cells, was not a good electron donor to the RC in the strain VC-F cells. When theB. viridis cytochrome c 2 was added to the membrane of the strain VC-F, re-reduction of the RC-bound cytochrome and concomitant oxidation of the cytochromec 2 were observed with a second order rate constant of 7.75 × 10−6, which was comparable with the value obtained from the measurement for the B. viridismembrane, 5.05 × 10−6. The membrane of the strain IL144RL2 did not show an apparent electron transfer reaction with theB. viridis cytochrome c 2. In this study, the RC-bound cytochrome subunit of B. viridis was shown to be synthesized in the distantly related species of purple bacteria, R. gelatinosus. Measurements of spectra and flash-induced redox changes of the cytochromes cshowed that the B. viridis cytochrome subunit was not only synthesized in R. gelatinosus but also functioned as the rapid electron donor to the photooxidized special pair. This means that the B. viridis cytochrome subunit can form a functional “chimeric” RC complex together with the R. gelatinosusLM core polypeptides. This was somewhat surprising, since the amino acid sequence identities of the L, M, and cytochrome subunits betweenB. viridis and R. gelatinosus are only 66, 61, and 45%, respectively. It has been reported in the fine structure of the B. viridis RC that 30 amino acids in the cytochrome subunit are bound to the LM core via hydrogen bonds and salt bridges (10Deisenhofer J. Epp O. Sinning I. Michel H. J. Mol. Biol. 1995; 246: 429-457Crossref PubMed Scopus (563) Google Scholar). These amino acids are indicated with the primary structure shown in Fig. 9. In the cytochrome subunit of R. gelatinosus, only 8 of these 30 amino acids are conserved, based on the comparison of the amino acid sequences. Some of the amino acid residues located near the binding residues are absent in the R. gelatinosus cytochrome subunit (Fig. 9). These observations suggest that the local protein structure contributing to the binding should be different between the cytochrome subunits of B. viridis and R. gelatinosus. It is known that the cytochrome subunit of R. gelatinosus is easily dissociated from the LM core by a moderate detergent treatment (36Fukushima A. Matsuura K. Shimada K. Satoh T. Biochim. Biophys. Acta. 1988; 933: 399-405Crossref Scopus (49) Google Scholar), although that of B. viridisis tightly bound to the core, possibly reflecting the structural difference. The cytochrome subunit of B. viridis seemed to be tightly bound to the R. gelatinosus LM core in the strain VC-F, since the RC complex with the cytochrome subunit was successfully isolated from the membranes of the strain VC-F under conditions in which the subunit was dissociated from the core complex in the R. gelatinosus wild type. On the other hand, most of the amino acid residues connecting the cytochrome subunit, based on the fine structure of the B. viridis RC (10Deisenhofer J. Epp O. Sinning I. Michel H. J. Mol. Biol. 1995; 246: 429-457Crossref PubMed Scopus (563) Google Scholar), were well conserved in the L (12 of 15 residues) and M (14 of 19 residues) subunits of R. gelatinosus (Fig. 9). The amino acid sequences around these residues also seemed to be well conserved. These findings suggest that the surfaces on the LM core that bind to the cytochrome subunit are similar in the two species and that the R. gelatinosus LM core can interact rather tightly with the B. viridis cytochrome subunit. In addition, the C-terminal residues of the M subunit may also contribute to the formation of the chimeric RC complex, since the C-terminal 34 amino acid residues after the 289th valine of the M subunit of the strain VC-F were derived from B. viridis. It has been suggested that the C terminus region of the M subunit is strongly associated with the cytochrome subunit in B. viridis as well as in other species (37Nitschke W. Dracheva S.M. Blankenship R.E. Madigan M.T. Bauer C.E. Anoxygenic Photosynthetic Bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands1995: 775-805Google Scholar). Thus, the cytochrome subunit of B. viridis in the strain VC-F would be closely associated with the “chimeric” M subunit. It is known that the N-terminal 20 amino acid residues in the precursor form of the cytochrome subunit of B. viridis function as a signal peptide and are removed possibly after secretion for the periplasmic space (5Weyer K.A. Lottspeich F. Gruenberg H. Lang F. Oesterhelt D. Michel H. EMBO J. 1987; 6: 2197-2202Crossref PubMed Google Scholar). The N-terminal first amino acid of the mature form of the subunit is a cysteine residue, which is modified by two fatty acid molecules (38Weyer K.A. Schafer W. Lottspeich F. Michel H. Biochemistry. 1987; 26: 2909-2914Crossref Scopus (69) Google Scholar). The sequence LVAGC at positions 17–21 in the precursor protein of B. viridis may be responsible for this modification and for the cleavage by signal peptidase II (38Weyer K.A. Schafer W. Lottspeich F. Michel H. Biochemistry. 1987; 26: 2909-2914Crossref Scopus (69) Google Scholar). The cytochrome subunit of R. gelatinosus also has a similar motif, LLAGC, at positions 19–23 (31Nagashima K.V.P. Matsuura K. Ohyama S. Shimada K. J. Biol. Chem. 1994; 269: 2477-2484Abstract Full Text PDF PubMed Google Scholar). The post-translational modification shown in B. viridis presumably occurs in the cells of R. gelatinosus. Evidence for this comes from the results of SDS-PAGE, which showed that the mobility of a band corresponding to the B. viridis cytochrome subunit synthesized in the R. gelatinosus strain VC-F was identical with that in the B. viridis membrane, as shown in Fig. 3. The redox midpoint potentials and the spectroscopic characteristics of four hemes in the cytochrome subunit of the strain VC-F were almost the same as those in the original B. viridis complex. This result suggests that the association with the R. gelatinosus LM core did not significantly affect the electrochemical properties of the hemes, except possibly for hemec559. Thus, the cytochrome subunit derived from the B. viridis gene probably holds its original structure and functions in the chimerical RC complex in the R. gelatinosusstrains VC-F. The results of flash-induced spectrophotometric measurements in the membrane preparation of the R. gelatinosus strain VC-F indicated that the cytochrome subunit derived from B. viridis functioned as an efficient electron donor to the photooxidized special pair in the strain VC-F. The rate of electron transfer from soluble electron carriers to the hemes in the cytochrome subunit in the strain VC-F was largely different from that in the wild-type strains of R. gelatinosus. This can be explained by the fact that the major electron donor to the cytochrome subunit is cytochromec 2 in B. viridis but HiPIP inR. gelatinosus. A previous study showed that the R. gelatinosus HiPIP is a very poor electron donor to theB. viridis RC (39Osyczka A. Nagashima K.V.P. Sogabe S. Miki K. Shimada K. Matsuura K. J. Biol. Chem. 2001; 276: 24108-24112Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar). Consistently, an addition of the R. gelatinosus HiPIP to the strain VC-F membrane had no detectable effects on the reduction of the cytochrome, as shown in Fig. 8. This may be one of the reasons that the growth rate of the strain VC-F is considerably lower than those of the wild-type strains under photosynthetic conditions. On the other hand, the B. viridiscytochrome c 2 worked as a good electron donor to the chimeric RC in the strain VC-F membrane, as observed for the native RC of B. viridis. The native structure of the B. viridis cytochrome subunit seems to be well maintained in theR. gelatinosus strain VC-F. Previous studies have shown that the cluster of acidic residues immediately surrounding the heme-1 of the RC-bound cytochrome subunit forms an electrostatically favorable binding site to cytochromes inR. gelatinosus (22Osyczka A. Nagashima K.V.P. Sogabe S. Miki K. Yoshida M. Shimada K. Matsuura K. Biochemistry. 1998; 37: 11732-11744Crossref PubMed Scopus (30) Google Scholar, 23Osyczka A. Nagashima K.V.P. Shimada K. Matsuura K. Biochemistry. 1999; 38: 2861-2865Crossref PubMed Scopus (24) Google Scholar, 24Osyczka A. Nagashima K.V.P. Sogabe S. Miki K. Shimada K. Matsuura K. Biochemistry. 1999; 38: 15779-15790Crossref PubMed Scopus (24) Google Scholar). A comparison of the three-dimensional structure of the cytochrome subunit solved inB. viridis and that predicted for R. gelatinosussuggested that these species are different in the distribution of the acidic residues around the heme-1. As part of a future study, it will be interesting to introduce site-specific mutations to the chimeric RC to clarify the interaction between the cytochrome subunit and the cytochrome c 2 of B. viridis. Further physiological and kinetic studies and additional site-directed mutagenesis on the cytochrome subunit of the strain VC-F should be useful for the clarification of the precise mechanisms of the electron transfer and of the physiological roles of the cytochrome subunit. A great number of studies have been conducted on the cytochrome subunit of B. viridis, resulting in a significant body of available structural and kinetic information. This information and the system established for R. gelatinosus in this study have made it easier than before to obtain further information about the relationships between the structure and function in the electron transfer protein.
Referência(s)