- Email: [email protected]
Evidence for a cytochrome f-Rieske protein subcomplex in the cytochrome b6f system from spinach chloroplasts
ARCHIVESOF BIOCHEMISTRYAND BIOPHYSICS Vol. 260, No. 1, January, pp. 408-415,1988
Evidence for a Cytochrome f-Rieske Protein Subcomplex Cytochrome b6f System from Spinach Chloroplasts MOHAMED EL-DEMERDASH, Technical
University
Berlin,
in the
JOHANN SALNIKOW,l AND JOACHIM VATER
Institute of Biochemistry and Molecular Biology, Franklinstrasse 1000 Berlin 10, West Germany
29,
Received June 23,198’7, and in revised form September 15,1987.
The cytochrome bGfcomplex of spinach chloroplasts was prepared with minor modification according to the method of E. Hurt and G. Hauska ((1981) Eur. J. Biochem. 117, 591-599) replacing, however, the final ultracentrifugation step by hydroxyapatite chromatography as suggested by M. F. Doyle and C.-A Yu ((1985) Biochem. Biophys. Res. Commun. 131, 700-706). The purified complex was partially dissociated by treatment with 4 M urea or 0.1% sodium dodecyl sulfate (SDS) in the absence of reducing agents. A binary subcomplex consisting of cytochrome f and the Rieske iron-sulfur protein was observed under these conditions by three different methods: (a) hydroxyapatite chromatography; (b) extraction with an isopropanol/water/trifluoroacetic acid mixture; and (c) gel filtration in the presence of low SDS concentrations. The subcomplex dissociated into its components by treatment with mercaptoethanol. These results suggest a close interaction of the cytochrome f with the Rieske protein involving SH groups which under reducing conditions leads to complete dissociation of the subcomplex. o 1988Academic Press,Inc.
The organization of cytochrome b-type complexes in electron transfer chains of mitochondria (l), chloroplasts (2, 3), and photosynthesizing bacteria (4) is strikingly similar. The chloroplast cytochrome bsf complex is primarily responsible for the electron transfer between photosysterns I and II (5-7). In its minimal functional form, the complex appears to consist of only four different polypeptides, cytochrome f (cl in mitochondria), cytochrome b6 (with two heme groups), the Rieske iron-sulfur protein, and an additional protein of 17-kDa molecular mass (8). The availability of highly purified preparations of this complex has permitted detailed studies of structure-function relationships. There is ample evidence that a major portion of cytochrome f including the heme group resides on the lumen side of the thylakoid membrane (9-12); amino 1To whom correspondence should be addressed. 0003-9861/88 $3.00 Copyright All rights
0 1988 by Academic Press, Inc. of reproduction in any form resewed.
acid sequence data (13, 14) suggest likewise a long, relatively polar N-terminal segment comprising the putative heme binding ligands which due to its relative hydrophilicity would be preferentially located at the inner thylakoid membrane surface, whereas the hydrophobic C-terminal portion points rather to membrane anchoring. A similar arrangement in the cytochrome bGf complex, i.e., preferential location at the inner surface, is proposed for the Rieske protein (12). In support of this model it was shown that both polypeptides, cytochrome f and the Rieske protein, are sensitive toward proteolysis in the total complex in contrast to cytochrome bs and the 17-kDa protein which appear to be protected (15). Further evidence for a membrane surface association comes from controlled proteolysis and surface labeling studies in the homologous cytochrome bq complex of mitochondria (16-19). Finally, exposure of cytochrome cl (20) as well as of the Rieske protein (21,22) 408
SPINACH
CHLOROPLAST
CYTOCHROME
at the inner membrane surface has been demonstrated by reaction with specific antibodies. Photosystem II oxidizes water and reduces plastoquinone (23). Oxidation of plastoquinol is probably mediated by the Rieske protein as indicated from antibody (24) and plastoquinone photoaffinity labeling studies (25). Cytochrome f is believed to accept electrons from the Rieske iron-sulfur protein and to transfer them to plastocyanin (5, 26-28). After extraction of the Rieske protein the reduction of cytochrome f is drastically inhibited (29). The electron acceptor plastocyanin is likewise located at the lumen side of the chloroplast membrane (30). From functional considerations close proximity of the catalytically active portions of cytochrome f, the Rieske protein, and plastocyanin exposed at the inner surface of the thylakoid membrane must be postulated, since their interaction is required in electron transport. The interaction between plastocyanin and cytochrome f has been studied kinetically (31-33); the functional linkage of these two polypeptides has been demonstrated for the isolated protein components (34) as well as in the total cytochrome b&unit (15). In contrast, there is no report so far about such an interaction between f and the Rieske iron-sulfur protein in the same complex. In the present communication we report evidence for such an interplay by the isolation of a stable binary complex by different physical methods. MATERIALS
Purification Complex
AND
METHODS
of the Cytochrome b6f
Spinach thylakoids were prepared as described by Hurt and Hauska (3). Thylakoid membranes were washed only once with 2 M NaBr and were solubilized with a solution of 40 mM MEGA-10,’ 0.5% sodium cholate, 0.4 M ammonium sulfate, 0.2 M sucrose in 20
’ Abbreviations used: MEGA-10, decanoyl-Nmethylglucamide; AcA, polyacrylamide-agarose; SDS, sodium dodecyl sulfate; TFA, trifluoroacetic acid; Tricine, N-tris(hydroxymethyl)methylglycine.
f-RIESKE
PROTEIN
SUBCOMPLEX
409
mM Tricine-NaOH, pH 8.0, at 4°C for 45 min. After centrifugation at 9000 rpm for 60 min in a DupontSorvall GS3 rotor the cytochrome bsf complex was precipitated from the yellow-green supternatant with ammonium sulfate at 45-55% saturation. The pellet was dissolved in 3-6 ml 30 mM Tris-succinate, pH 6.5, containing 40 mM MEGA-10, 0.5% sodium cholate, and was stored overnight at 4°C. The solution of the complex was desalted on a Sephadex G-25 column (2 X 90 cm) which had been equilibrated with 50 mM potassium phosphate, pH 6.8, and 1% sodium cholate. The dark green fractions collected from this step were applied to a hydroxyapatite column (3 X 12 cm) equilibrated with the same buffer and eluted with 0.2 M potassium phosphate, pH 6.8, containing 1% sodium cholate. The cytochrome bsf complex was analyzed by SDS-polyacrylamide gel electrophoresis.
Dissociation of the Cytochrome b6f Complex with Urea in the Absence of Reducing Agents Partial dissociation of the cytochrome b6f complex prepared as described above was achieved by treatment with 4 M urea in a solution containing 30 mM Tris-succinate, pH 6.5, 0.5% sodium cholate, and 40 mM MEGAin the absence of reducing agents. The samples were incubated for 15 min with stirring and were desalted on a Sephadex G-25 column equilipH 6.5, containing brated with 30 mM Tris-succinate, 1% sodium cholate.
Isolation of the Cytochrome f-Rieske Protein Subcomplex under Nonreducing Conditions (a) Hydroxyapatite Chromatography. A sample of cytochrome b6fcomplex which was treated with urea and desalted as outlined above was loaded onto a hydroxyapatite column (1.5 X 10 cm) equilibrated with 10 mM potassium phosphate, pH 6.8, containing 1% sodium cholate. The column was eluted with the same buffer with the phosphate concentration raised t0
50 mM.
(b) Solvent extractiun. Samples of dissociated cytochrome b,f complex were exhaustively dialyzed against distilled water for 24 h at 4”C, then mixed with 4 vol of cold acetone (-20°C) and kept for 15 min at the same temperature. The pellet was removed by centrifugation, washed twice with distilled water, and extracted with a mixture of isopropanol:water:TFA (50:50:0.1, by vol). The extract was evaporated under reduced pressure, resuspended in 1 ml water, and lyophilized. (c) Gel fifiltration. A sample of purified cytochrome b6f complex was applied to an AcA-54 column (2
410
EL-DEMERDASH,
SALNIKOW,
X 100 cm) equilibrated with 20 mM Trieine-NaOH, pH 8.0, containing 0.1% SDS and eluted at a flow rate of 0.3 ml/min. Fractions of 1.5 ml were collected. The protein pattern resulting from each separation technique was analyzed by SDS-polyacrylamide gel electrophoresis.
Eflect of Mercaptoethanol on the Dissociation of the Cytochrome b6f Complex as Analyzed by Solvent Extraction and Gel Filtration The purified complex was dialyzed, lyophilized, and dissolved in 4 M urea containing 1% sodium cholate, 5 mM EDTA, and 0.5 M Tris-HCI, pH 8.5. Aliquots were treated with increasing concentrations of mercaptoethanol (0.01, 0.05, 0.1, 0.2, 0.3., and 0.4 mM) for 6 h under a nitrogen atmosphere. After reduction, sulfhydryl groups were aminoethylated by the addition of three portions of ethyleneimine at lo-min intervals. A molar ratio of mercaptoethanol:ethyleneimine of 1:3 was used. Reduced and aminoethylated cytochrome b&complexes were separated from salts and reagents by dialysis against distilled water overnight. The dialysates were centrifuged for 15 min at 10,000 rpm and the supernatants were discarded. The pellets were extracted twice with 5 ml each of isopropanol:water:TFA (50:50:0.1, by vol) followed by centrifugation for 10 min at 10,000 rpm. The supernatants were evaporated in oacuo, lyophilized, and analyzed by SDS-polyacrylamide gel electrophoresis. A sample of cytochrome &fcomplex was subjected to gel filtration on AcA-54 in 20 mM Tricine-NaOH, pH 8.0, containing 0.1% SDS as described above, however, mercaptoethanol was added to the sample and the eluant at a concentration of 0.5% (v/v).
Analytical
Methods
Protein was determined by the method of Lowry et al. (35). SDS-polyacrylamide gel electrophoresis was performed according to Laemmli (36).
AND
VATER
many). The column supports, hydroxyapatite (Hypatite C), and AcA-54 were purchased from LKB and Sephadex G-25 was from Pharmacia. RESULTS
Dissociaticm of the Cytochrome bsf Complex under Now-educing Conditions
The cytochrome bsf complex as isolated from spinach thylakoids by ammonium sulfate fractionation and hydroxyapatite chromatography shows a polypeptide composition of four bands (Fig. 1) as reported by several authors: cytochrome f with a molecular mass of 33 to 34 kDa, cytochrome bs of 23 kDa, the Rieske ironsulfur protein of 20 kDa, and an additional component of 17 kDa with yet undefined function. Treatment of this complex with 4 M urea at pH 6.5 for 15 min in the absence of mercaptoethanol and hydroxyapatite chromatography yields a polypeptide pattern containing cytochrome f and the Rieske protein as the most prominent protein bands (Fig. 2a). A cytochrome fRieske subcomplex is also obtained by solvent extraction with 50% isopropanol containing 0.1% TFA. The sample demonstrated in Fig. 2b still shows traces of the 17-kDa component. Since it could be ar-
Molecular Weight ( KDa)
94 67 43 30
Materials Spinach was obtained from the local market; leaves were washed, deveined, and stored overnight at 4°C before use. Sodium cholate and n-octyl-glucoside (1-0-n-octyl-@-D-glucopyranoside) were from Sigma (Deisenhofen, Germany) and MEGA(decanoyl-N-methylglucamide) was from Oxyl GmbH (Bobingen, Germany). Reagents for gel electrophoresis were purchased from Serva (Heidelberg, Germany) and LKB (Grafelfing, Germany). Standard proteins for the SDS-polyacrylamide gel electrophoresis were obtained from Pharmacia (Freiburg, Ger-
1
2
3
L
FIG. 1. SDS-polyacrylamide gel electrophoresis (15%) of the cytochrome b&complex after hydroxyapatite chromatography. Marker proteins: Phosphorylase (94 kDa); bovine serum albumin (67 kDa); ovalbumin (43 kDa); carbonic anhydrase (30 kDa); soybean trypsin inhibitor (20.1 kDa); and a-lactalbumin (14.4 kDa).
SPINACH
CHLOROPLAST
CYTOCHROME
f-RIESKE
PROTEIN
SUBCOMPLEX
411
Mol;:i”d~~
a
(KDa)
12
b
3
12
C
3
123456
FIG. 2. SDS-polyacrylamide gel electrophoresis (15%). (a) Cytochromef-Rieske protein complex as obtained from cytochrome b,Jcomplex by partial dissociation with 4 M urea and hydroxyapatite chromatography. Lane 1, marker proteins; lanes 2 and 3, cytochromef-Rieske protein subcomplex. (b) Cytochrome f-Rieske protein subcomplex as obtained by extraction of the cytochrome b6f complex with 50% isopropanol containing 0.1% TFA in the absence of reducing agents. Lane 1, cytochrome bsf complex prepared according to Doyle and Yu (37); lane 2, cytochrome b6f complex purified according to Hurt and Hauska (3); lane 3, cytochrome f-Rieske protein subcomplex. (c) Rieske iron-sulfur protein obtained by solvent extraction of a cytochrome b6f complex which had been reduced with increasing concentrations of mercaptoethanol and aminoethylated. The purified complex was dialyzed, lyophilized, and dissolved in 0.5 M Tris-HCI buffer, pH 8.5, containing 4 M urea, 1% sodium cholate, and 5 mM EDTA, as indicated under Materials and Methods. The final concentration of the complex was equivalent to approximately 10 nmol Cytf/ml. Aliquots of 2 ml were reduced with mercaptoethanol and extracted with 50% isopropanol containing 0.1% TFA. Sequential increased enrichment of the Rieske protein as a function of the reductant concentration was detected by SDS-polyacrylamide gel electrophoresis. Lanes l-6: 0.01, 0.05, 0.1, 0.2, 0.3, and 0.4 mM mercaptoethanol, respectively.
gued that these two polypeptides are perhaps separated rather accidentally because of their similar physical properties rather than subcomplex formation, dissociation of the cytochrome b6f complex by 0.1% SDS and separation by molecular mass using gel filtration in the presence of detergent was attempted. Figure 3a shows that the assumed cytochrome f-Rieske protein subcomplex possessing the highest molecular mass elutes at the front near the void volume, is followed by a second putative binary subcomplex consisting of cytochrome b6 and the 17-kDa protein, and finally by unbound cytochrome b6. Effect of Mercaptoethanol Formation
on Subcomplex
Since the cytochrome b,f complex participates in plants in the electron flow of both photosystems, the effect of reducing
agents like mercaptoethanol on the subcomplex formation appeared interesting. Cytochrome b6f complexes were treated with various concentrations of mercaptoethanol and reacted with ethyleneimine in order to stabilize the reduced state by chemical modification. Samples of thus modified complexes were again extracted with 50% isopropanol (containing 0.1% TFA) and analyzed by SDS-polyacrylamide gel electrophoresis (Fig. 2~). The presence of mercaptoethanol leads to a selective extraction of almost pure Rieske iron-sulfur protein with only traces of cytochrome f present. Figure 2c shows sequential increased enrichment of the FeS protein at increasing concentrations of the reductant. An analogous experiment involving gel filtration of the cytochrome b&complex in 0.1% SDS in the presence of mercaptoethanol was performed (Fig. 3b). In contrast
412
EL-DEMERDASH,
SALNIKOW,
AND
VATER
Cyt f------...----CYt b6 --___---__
Rieske
-FeS-
Direction
of
elution.-w
FIG. 3. SDS-polyacrylamide eleetrophoresis (15%) of fractions obtained from gel filtration of the cytochrome &fcomplex on AcA-54 in the presence of 0.1% SDS (a) under nonreducing conditions, or (b) in the presence of 0.5% mercaptoethanol.
to the subcomplex pattern observed under nonreducing conditions the four polypeptides now eluted in order of their molecular masses. These results indicate that the binary complex obtained in the absence of mercaptoethanol represents a native building block of the cytochrome b,f architecture stabilized by molecular forces which are sensitive toward reducing agents. DISCUSSION
In this study the cytochrome b,f complex from spinach chloroplast was prepared according to the method of Hurt and Hauska (3), however, the time-consuming final ultracentrifugation step was replaced by hydroxyapatite chromatography as recommended by Doyle and Yu (37) yielding considerable gains in time and sample capacity. Recently, it was shown that by using this method active cytochrome b&complexes could be prepared on a large scale in at least 90% pure form
(38). The polypeptide composition of our preparation is in close agreement with other published data (3,37-39). The dissociation studies in the cytochrome b6f system using mild denaturants such as 4 M urea or 0.1% SDS as described in this report point to a subarchitecture of the entire complex consisting of the binary building blocks cytochrome fRieske protein and cytochrome bs-1’7-kDa protein (= subunit 4 of the cytochrome bsf complex). A functional linkage of cytochrome f and the Rieske iron-sulfur protein has been deduced by several authors (5,26-28); however, isolation of a binary molecular complex proving the molecular proximity of both components has to our knowledge been demonstrated here for the first time. Thus, the cytochrome f-Rieske protein complex represents another link in the chain of adjacent electron carriers for which interaction has been demonstrated such as cytochrome f/plastocyanin (31-34) or ferredoxin/ferredoxin-NADP+ reductase pairs (40-43). In
SPINACH
CHLOROPLAST
CYTOCHROME
mitochondria, the Rieske iron-sulfur protein is closely associated with cytochrome cl which is functionally the biological equivalent of cytochrome f in this organelle. It has been shown recently that both components interact tightly in the complex III of beef heart mitochondria (44) thus constituting an integral part of the electron pathway from ubiquinol to cytochrome c. In view of the evolutionary resemblance of the chloroplasts and mitochondrial organelles-specifically the electron carrier systems involved in energy transduction-it is not surprising that the molecular environment elucidated for the mitochondrial Rieske ironsulfur protein is mirrored to some extent in the chloroplast system. A second observation evident from the gel filtration experiment is a close association of cytochrome b6 with the l’i-kDa protein. Again, bearing in mind the structural similarities between the mitochondrial complex III and the chloroplast b6f complex, this result is not unexpected, since the l’l-kDa protein shows considerable sequence homology to the C-terminal end of mitochondrial cytochrome b, thus pointing to a common ancestor gene early in the evolution (45). The forces involved in the interaction of cytochrome f and the Rieske iron-sulfur protein are at present not clear. Since complete dissociation of the cytochrome b&complex in the presence of detergents requires reducing conditions as, for example, mercaptoethanol, apparently not only hydrophobic interactions are responsible for the stability of the subcomplex, but disulfide bridges seem to play an important role as well. However, taking into account available sequence data and present concepts of thylakoid membrane organization a direct disulfide linkage between cytochrome f and the Rieske iron-sulfur protein appears difficult to formulate: the major N-terminal part of cytochrome f which is exposed to the lumen of the thylakoid membrane possesses only two cysteine residues (46) which are presumed to serve as heme attachment sites thus eliminating possible additional thiol linkages. The Rieske protein of plant chloroplasts,
f-RIESKE
PROTEIN
SUBCOMPLEX
413
on the other hand, as analyzed by sequencing of the 96 N-terminal residues and isolation of additional peptide fragments possesses five cysteinyl residues (47): four cysteins are found in conserved sequences which match with those determined in the C-termini of Rieske proteins from fungal and bacterial cytochrome bcl complexes (48,49) and which are thought to represent the (2Fe-2s) cluster binding sites; an additional cysteinyl peptide has been identified which apparently does not participate in iron-sulfur core binding (48). The requirement for four cysteines in iron binding, however, is not yet certain in the chloroplast system; recent studies on the respiratory chain of thermophilic bacteria point to nitrogen atoms as additional coordination sites for iron (50, 51) thus limiting the number of essential cysteinyl side chain functions for cluster formation to two. Consequently, the chloroplast Rieske iron-sulfur protein possesses one to possibly three potential thiol functions qualifying for disulfide bridge formation. The observation of a cytochrome fRieske protein subcomplex necessitates a discussion of the structural organization of the cytochrome b6f complex in the thylakoid membrane, although solid data are scarce here and the models are somewhat controversial compared to the analogous mitochondrial cytochrome beI systems. There is ample evidence that in the cytochrome beI complex the Rieske iron-sulfur protein and cytochrome cl traverse the membrane displaying their functional groups in close proximity (18, 44); the amino acid sequences of the Rieske proteins from the cytochrome bq complex of Neurospora and Rhodopseudomonas sphaeroides (48, 49) show a hydrophobic segment at the N-terminal half long enough to span the membrane barrier anchoring the protein in it; the iron-sulfur cluster linked to the more hydrophilic Cterminal portion would be exposed into the intermembrane space on the inner membrane surface. Sequence data of the chloroplast Rieske protein (47) point to an analogous membrane orientation involving N-terminal anchoring in the membrane via a hydrophobic membrane span-
414
EL-DEMERDASH,
SALNIKOW,
ning segment. The iron-sulfur cluster as the prosthetic group of the catalytic domain would be located at the C-terminal part and, consequently, exposed toward the thylakoid lumen, thus placing the Nterminus toward the stromal surface. This orientation would permit close contact of the iron-sulfur cluster of the Rieske protein with the similarly oriented heme function of cytochrome $ In support of this orientation antibodies against the Rieske protein have been shown to not inhibit electron transport from water or plastoquinol to photosystem I in chloroplasts (24). In contrast to this model, studies using limited proteolysis and specific antibodies favor an opposite orientation of the Rieske protein with the putative ironsulfur cluster carrying the C-terminal fragment close to or on the stromal side of the thylakoid membrane surface (52). This model, however, places severe spatial constraints on the functional heme-iron-sulfur interaction required for electron flow. The dissociating effect of reducing agents such as mercaptoethanol could perhaps be a function of the iron-sulfur core of the Rieske protein itself. Under dissociating conditions such as with 0.1% SDS the iron-sulfur cluster is fully retained as has been demonstrated by Nishikimi et al. (53) permitting subsequent reactions with redox reagents. Furthermore, it has been shown that the Rieske protein is more readily extracted from the mitochondrial complex III under reducing conditions (54) pointing again to an involvement of thiol groups in binding. In the cytochrome bq system a tenacious association of the Rieske protein with cytochrome cl has been observed; final separation was achieved by passage of the partially purified system through a mercurial resin column which retained the cytochrome component, however, not the Rieske protein (55), indicating that the latter does not possess free SH groups on its surface when complexed to cytochrome cl. On the other hand, cleavage of the total cytochrome reductase complex in Triton X-100 by higher salt concentrations in the presence of ascorbate and dithioerythritol leads to facile dissociation yielding the
AND
VATER
iron-sulfur subunit and complexes lacking redox centers as separable components (56). Considering the structural and functional similarities of the cytochrome bq and cytochrome b6f systems, these observations appear as additional support for the experimental data described in this report. In summary, the available information on the structure of the cytochrome b6f complex and the stability of the cytochrome f-Rieske protein subcomplex can perhaps be explained by an intramolecular disulfide bridge of the Rieske protein itself stabilizing a conformation required for tight interaction of the heme and iron-sulfur chromophores; this would be in accordance with the structural and three-dimensional features of current models. In view of a lack of additional supportive evidence, however, this hypothesis remains largely speculative at present. REFERENCES 1. TRUMPOWER, B. L., AND KATKI, A. G. (1979) in Membrane Proteins in Energy Transduction (Capaldi, R. A., Ed.), pp. 89-200, Dekker, Basel/New York. 2. NELSON, N., AND NEUMANN, J. (1972) J. Biol. Chem. 247,1917-1924. 3. HURT, E., AND HAUSKA, G. (1981) Eur. J. Biothem. 117,591-599. 4. CROFTS, A. R., MEINHARDT, S. W., AND BOWYER, J. R. (1981) in Function of Quinones in Energy Conversing Systems (Trumpower, B. L., Ed.), Academic Press, New York. 5. HAUSKA, G., HURT, E., GABELLINI, N., AND LOCKAU, W. (1983) Biochim Biophys. Acta 726, 97-133. 6. BARBER, J. (1984) Trends B&hem Sci. 9,209-210. 7. HAEHNEL, W. (1984) Annu. Rev. Plant. Physiol 35,659-693. 8. HURT, E., AND HAUSKA, G. (1982) Biochim. Biophys. Acta 682,466-473. 9. MORSCHEL, E., AND STAEHELIN, L. A. (1983) J. Cell Biol. 97, 301-310. 10. WILLEY, D. L., AUFFRET, A. D., AND GRAY, J. C. (1984) Cell 36,555-562. 11. MANSFIELD, R. W., AND BENDALL, D. S. (1984) B&him Biophys. Acta 766,62-69. 12. MANSFIELD, R. W., AND ANDERSON, J. M. (1985) Biochim. Biophys. Acta 809.435-444. 13. WILLEY, D. L., AND GARY, J. C. (1984) in Advances in Photosynthesis Research (Sybesma,
SPINACH
14.
15. 16. 17.
18. 19.
20.
21.
22. 23. 24. 25.
26. 27.
28.
29. 30. 31.
CHLOROPLAST
CYTOCHROME
C., Ed.), Vol. IV, pp. 567-570, Nijhoff, Junk, The Hague. ALAM, J., AND KROGMANN, I. W. (1984) in Advances in Photosynthesis Research (Sybesma, C., Ed.), Vol. I, pp. 521-524, Nijhoff, Junk, The Hague. HURT, E. (1983) Dissertation, University of Regensburg. GELLERFORS, P., AND NELSON, B. D. (1978) FEBS Lett. 92.36-40. BELL, R. L., SWEETLAND, J., LUDWIG, B., AND CAPALDI, R. A. (1979) Proc. N&l. Acad. Sci. USA 76,741-745. LI, Y., DE VRIES, S., LEONARD, K., AND WEISS, H. (1981) FEBS Lett. 135,277-280. LEONARD, K., HOVM~LLER, S., KARLSSON, B., WINGFIELD, P., LI, Y., PERKINS, S., AND WEISS, H. (1981) in Vectorial Research in Electron and Ion Transport in Mitochondria and Bacteria (Palmieri, F., Quagliarello, E., Siliprandi, N., and Slater, E. C., Eds.), pp. 129-137, Elsevier/North-Holland, Amsterdam. SHNIEDER, D. L., AND RACKER, E. (1973) in Oxidases and Related Redox Systems (King, T. E., Mason, H. S., and Marrison, M., Eds.), pp. 799-821, Wiley, New York. BEATTIE, D. S., SIDHU, A., AND CLEJAN, L. (1982) in Proceedings, 2nd European Bioengineering Conference Short report, pp. 203-204, University Claude Beranard, Lyon. SIDHU, A., AND BEATTIE, D. S. (1982) J. Biol. Chem. 257,7879-7886. Biophys. BOUGES-BOCQUET, B. (1980) Biochim. Acta 594, 85-103. HURT, E., HAUSKA, G., AND MALKIN, R. (1981) FEBS Lett. 134, l-5. OETTMEIER, W., MASSON, K., SOLL, H.-J., HURT, E., AND HAUSKA, G. (1982) FEBS Lett. 144, 313-317. VELTHUYS, B. R. (1980) Annu. Rev. Plant Physiol. 31,545-567. MALKIN, R. (1982) in Electron Transport and Phosphorylation (Barber, J., Ed.), Vol. 4, pp. 2-47, Elsevier Biochemical, Amsterdam/New York. CRAMER, W. A., AND CROFTS, A. R. (1982) in Photosynthesis (Govindjee, Ed.), Vol. I, pp. 387-467, Academic Press, New York. LAM, E., AND MALKIN, R. (1982) FEBS Lett. 144, 190-193. HAEHNEL, W., BERZBORN, R. J., AND ANDERSSON, B. (1981) B&him. Biophys. Acta 637,389-399. WOOD, P. M. (1974) Biochim. Biophys. Acta 357, 370-379.
f-RIESKE
PROTEIN
SUBCOMPLEX
415
32. NIWA, S., ISCHIKAWA, H., NIKAI, S., AND TABAKE, T. (1980) J. Biol. Chem. 88,1177-1183. 33. TANAKE, K., TAKAHASHI, M., AND ASADA, K. (1981) Plant Cell PhysioL 22,33-39. 34. DAVIS, D. J., AND HOUGH, K. (1983) Biochem. Biophys. Res. Cammun. 116,1000-1006. 35. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. Biol. Chem. 193, 265-275. 36. LAEMMLI, U. K. (1979) Nature (London) 227, 680-685. 37. DOYLE, M. F., AND Yu, C.-A. (1985) Biochem. Biophys. Res. Commun. 131,700-706. 38. BLACK, M. T., WIDGER, W. R., AND CRAMER, W. A. (1987) Arch. Biochem. Biophys. 252,655-661. 39. CLARK, R. D., AND HIND, G. (1983) J. Biol. Chem. 258,10348-10354. 40. SHIN, M., AND SAN PIETRO, A. (1968) Biochem. Biophys. Res. Cwmmun. 33,38-42. 41. FOUST, G. P., MAYHEW, S. G., AND MASSEY, V. (1969) J. Biol. Chem. 244,964-970. 42. BAGER, P. (1971) Planta 99,319-338. 43. SHIN, M. (1973) Biochim Biophys. Acta 292, 13-19. 44. SHIMOMURA, Y., NISHIKIMI, M., AND OZAWA, T. (1985) J. BioZ. Chem. 260,15075-15080. 45. WIDGER, W. R., CRAMER, W. A., HERRMANN, R. G., AND TREBST, A. (1984) Proc. Natl. Acad Sci. USA 81,674-678. 46. ALT, J., AND HERRMANN, R. G. (1984) Curr. Genet. 8,551-557. 47. PFEFFERKORN, B., AND MEYER, H. E. (1986) FEBS Lett. 206,233-237. 48. HARNISCH, K., WEISS, H., AND SEBALD, W. (1985) Eur. J. B&hem. 149,95-99. 49. GABELLINI, N., AND SEBALD, W. (1986) Eur. J. Biochem. 154,569-579. 50. FEE, J. A., FINDLING, K. L., YOSHIDA, T., HILLE, R., TARR, G. E., HEARCHEN, D. O., DUNHAM, W. R., DAY, E. P., KENT, T. A., AND M~NCK, E. (1984) J. Biol. Chem. 259,124-133. 51. KUILA, D., AND FEE, J. A. (1986) J. Biol. Chem. 261,2768-2771. 52. MANSFIELD, R. W., AND ANDERSON, J. M. (1985) Biochim. Biophys. Acta 809,435-444. 53. NISHIKIMI, M., SHIMOMURA, Y., YAMADA, H., AND OZAWA, T. (1984) Biochem. Bioph,ys. Res. Cammm. 120,237-241. 54. RIESKE, J. S. (1976) Biochim. Biophys. Acta 456, 195-247. 55. TRUMPOWER, B. L., AND EDWARDS, C. A. (1979) J. Biol. Chem. 254,8697-8706. 56. HOVM~LLER, S., LEONARD, K., AND WEISS, H. (1981) FEBS Lett. 123,118-122.
Evidence for a Cytochrome f-Rieske Protein Subcomplex Cytochrome b6f System from Spinach Chloroplasts MOHAMED EL-DEMERDASH, Technical
University
Berlin,
in the
JOHANN SALNIKOW,l AND JOACHIM VATER
Institute of Biochemistry and Molecular Biology, Franklinstrasse 1000 Berlin 10, West Germany
29,
Received June 23,198’7, and in revised form September 15,1987.
The cytochrome bGfcomplex of spinach chloroplasts was prepared with minor modification according to the method of E. Hurt and G. Hauska ((1981) Eur. J. Biochem. 117, 591-599) replacing, however, the final ultracentrifugation step by hydroxyapatite chromatography as suggested by M. F. Doyle and C.-A Yu ((1985) Biochem. Biophys. Res. Commun. 131, 700-706). The purified complex was partially dissociated by treatment with 4 M urea or 0.1% sodium dodecyl sulfate (SDS) in the absence of reducing agents. A binary subcomplex consisting of cytochrome f and the Rieske iron-sulfur protein was observed under these conditions by three different methods: (a) hydroxyapatite chromatography; (b) extraction with an isopropanol/water/trifluoroacetic acid mixture; and (c) gel filtration in the presence of low SDS concentrations. The subcomplex dissociated into its components by treatment with mercaptoethanol. These results suggest a close interaction of the cytochrome f with the Rieske protein involving SH groups which under reducing conditions leads to complete dissociation of the subcomplex. o 1988Academic Press,Inc.
The organization of cytochrome b-type complexes in electron transfer chains of mitochondria (l), chloroplasts (2, 3), and photosynthesizing bacteria (4) is strikingly similar. The chloroplast cytochrome bsf complex is primarily responsible for the electron transfer between photosysterns I and II (5-7). In its minimal functional form, the complex appears to consist of only four different polypeptides, cytochrome f (cl in mitochondria), cytochrome b6 (with two heme groups), the Rieske iron-sulfur protein, and an additional protein of 17-kDa molecular mass (8). The availability of highly purified preparations of this complex has permitted detailed studies of structure-function relationships. There is ample evidence that a major portion of cytochrome f including the heme group resides on the lumen side of the thylakoid membrane (9-12); amino 1To whom correspondence should be addressed. 0003-9861/88 $3.00 Copyright All rights
0 1988 by Academic Press, Inc. of reproduction in any form resewed.
acid sequence data (13, 14) suggest likewise a long, relatively polar N-terminal segment comprising the putative heme binding ligands which due to its relative hydrophilicity would be preferentially located at the inner thylakoid membrane surface, whereas the hydrophobic C-terminal portion points rather to membrane anchoring. A similar arrangement in the cytochrome bGf complex, i.e., preferential location at the inner surface, is proposed for the Rieske protein (12). In support of this model it was shown that both polypeptides, cytochrome f and the Rieske protein, are sensitive toward proteolysis in the total complex in contrast to cytochrome bs and the 17-kDa protein which appear to be protected (15). Further evidence for a membrane surface association comes from controlled proteolysis and surface labeling studies in the homologous cytochrome bq complex of mitochondria (16-19). Finally, exposure of cytochrome cl (20) as well as of the Rieske protein (21,22) 408
SPINACH
CHLOROPLAST
CYTOCHROME
at the inner membrane surface has been demonstrated by reaction with specific antibodies. Photosystem II oxidizes water and reduces plastoquinone (23). Oxidation of plastoquinol is probably mediated by the Rieske protein as indicated from antibody (24) and plastoquinone photoaffinity labeling studies (25). Cytochrome f is believed to accept electrons from the Rieske iron-sulfur protein and to transfer them to plastocyanin (5, 26-28). After extraction of the Rieske protein the reduction of cytochrome f is drastically inhibited (29). The electron acceptor plastocyanin is likewise located at the lumen side of the chloroplast membrane (30). From functional considerations close proximity of the catalytically active portions of cytochrome f, the Rieske protein, and plastocyanin exposed at the inner surface of the thylakoid membrane must be postulated, since their interaction is required in electron transport. The interaction between plastocyanin and cytochrome f has been studied kinetically (31-33); the functional linkage of these two polypeptides has been demonstrated for the isolated protein components (34) as well as in the total cytochrome b&unit (15). In contrast, there is no report so far about such an interaction between f and the Rieske iron-sulfur protein in the same complex. In the present communication we report evidence for such an interplay by the isolation of a stable binary complex by different physical methods. MATERIALS
Purification Complex
AND
METHODS
of the Cytochrome b6f
Spinach thylakoids were prepared as described by Hurt and Hauska (3). Thylakoid membranes were washed only once with 2 M NaBr and were solubilized with a solution of 40 mM MEGA-10,’ 0.5% sodium cholate, 0.4 M ammonium sulfate, 0.2 M sucrose in 20
’ Abbreviations used: MEGA-10, decanoyl-Nmethylglucamide; AcA, polyacrylamide-agarose; SDS, sodium dodecyl sulfate; TFA, trifluoroacetic acid; Tricine, N-tris(hydroxymethyl)methylglycine.
f-RIESKE
PROTEIN
SUBCOMPLEX
409
mM Tricine-NaOH, pH 8.0, at 4°C for 45 min. After centrifugation at 9000 rpm for 60 min in a DupontSorvall GS3 rotor the cytochrome bsf complex was precipitated from the yellow-green supternatant with ammonium sulfate at 45-55% saturation. The pellet was dissolved in 3-6 ml 30 mM Tris-succinate, pH 6.5, containing 40 mM MEGA-10, 0.5% sodium cholate, and was stored overnight at 4°C. The solution of the complex was desalted on a Sephadex G-25 column (2 X 90 cm) which had been equilibrated with 50 mM potassium phosphate, pH 6.8, and 1% sodium cholate. The dark green fractions collected from this step were applied to a hydroxyapatite column (3 X 12 cm) equilibrated with the same buffer and eluted with 0.2 M potassium phosphate, pH 6.8, containing 1% sodium cholate. The cytochrome bsf complex was analyzed by SDS-polyacrylamide gel electrophoresis.
Dissociation of the Cytochrome b6f Complex with Urea in the Absence of Reducing Agents Partial dissociation of the cytochrome b6f complex prepared as described above was achieved by treatment with 4 M urea in a solution containing 30 mM Tris-succinate, pH 6.5, 0.5% sodium cholate, and 40 mM MEGAin the absence of reducing agents. The samples were incubated for 15 min with stirring and were desalted on a Sephadex G-25 column equilipH 6.5, containing brated with 30 mM Tris-succinate, 1% sodium cholate.
Isolation of the Cytochrome f-Rieske Protein Subcomplex under Nonreducing Conditions (a) Hydroxyapatite Chromatography. A sample of cytochrome b6fcomplex which was treated with urea and desalted as outlined above was loaded onto a hydroxyapatite column (1.5 X 10 cm) equilibrated with 10 mM potassium phosphate, pH 6.8, containing 1% sodium cholate. The column was eluted with the same buffer with the phosphate concentration raised t0
50 mM.
(b) Solvent extractiun. Samples of dissociated cytochrome b,f complex were exhaustively dialyzed against distilled water for 24 h at 4”C, then mixed with 4 vol of cold acetone (-20°C) and kept for 15 min at the same temperature. The pellet was removed by centrifugation, washed twice with distilled water, and extracted with a mixture of isopropanol:water:TFA (50:50:0.1, by vol). The extract was evaporated under reduced pressure, resuspended in 1 ml water, and lyophilized. (c) Gel fifiltration. A sample of purified cytochrome b6f complex was applied to an AcA-54 column (2
410
EL-DEMERDASH,
SALNIKOW,
X 100 cm) equilibrated with 20 mM Trieine-NaOH, pH 8.0, containing 0.1% SDS and eluted at a flow rate of 0.3 ml/min. Fractions of 1.5 ml were collected. The protein pattern resulting from each separation technique was analyzed by SDS-polyacrylamide gel electrophoresis.
Eflect of Mercaptoethanol on the Dissociation of the Cytochrome b6f Complex as Analyzed by Solvent Extraction and Gel Filtration The purified complex was dialyzed, lyophilized, and dissolved in 4 M urea containing 1% sodium cholate, 5 mM EDTA, and 0.5 M Tris-HCI, pH 8.5. Aliquots were treated with increasing concentrations of mercaptoethanol (0.01, 0.05, 0.1, 0.2, 0.3., and 0.4 mM) for 6 h under a nitrogen atmosphere. After reduction, sulfhydryl groups were aminoethylated by the addition of three portions of ethyleneimine at lo-min intervals. A molar ratio of mercaptoethanol:ethyleneimine of 1:3 was used. Reduced and aminoethylated cytochrome b&complexes were separated from salts and reagents by dialysis against distilled water overnight. The dialysates were centrifuged for 15 min at 10,000 rpm and the supernatants were discarded. The pellets were extracted twice with 5 ml each of isopropanol:water:TFA (50:50:0.1, by vol) followed by centrifugation for 10 min at 10,000 rpm. The supernatants were evaporated in oacuo, lyophilized, and analyzed by SDS-polyacrylamide gel electrophoresis. A sample of cytochrome &fcomplex was subjected to gel filtration on AcA-54 in 20 mM Tricine-NaOH, pH 8.0, containing 0.1% SDS as described above, however, mercaptoethanol was added to the sample and the eluant at a concentration of 0.5% (v/v).
Analytical
Methods
Protein was determined by the method of Lowry et al. (35). SDS-polyacrylamide gel electrophoresis was performed according to Laemmli (36).
AND
VATER
many). The column supports, hydroxyapatite (Hypatite C), and AcA-54 were purchased from LKB and Sephadex G-25 was from Pharmacia. RESULTS
Dissociaticm of the Cytochrome bsf Complex under Now-educing Conditions
The cytochrome bsf complex as isolated from spinach thylakoids by ammonium sulfate fractionation and hydroxyapatite chromatography shows a polypeptide composition of four bands (Fig. 1) as reported by several authors: cytochrome f with a molecular mass of 33 to 34 kDa, cytochrome bs of 23 kDa, the Rieske ironsulfur protein of 20 kDa, and an additional component of 17 kDa with yet undefined function. Treatment of this complex with 4 M urea at pH 6.5 for 15 min in the absence of mercaptoethanol and hydroxyapatite chromatography yields a polypeptide pattern containing cytochrome f and the Rieske protein as the most prominent protein bands (Fig. 2a). A cytochrome fRieske subcomplex is also obtained by solvent extraction with 50% isopropanol containing 0.1% TFA. The sample demonstrated in Fig. 2b still shows traces of the 17-kDa component. Since it could be ar-
Molecular Weight ( KDa)
94 67 43 30
Materials Spinach was obtained from the local market; leaves were washed, deveined, and stored overnight at 4°C before use. Sodium cholate and n-octyl-glucoside (1-0-n-octyl-@-D-glucopyranoside) were from Sigma (Deisenhofen, Germany) and MEGA(decanoyl-N-methylglucamide) was from Oxyl GmbH (Bobingen, Germany). Reagents for gel electrophoresis were purchased from Serva (Heidelberg, Germany) and LKB (Grafelfing, Germany). Standard proteins for the SDS-polyacrylamide gel electrophoresis were obtained from Pharmacia (Freiburg, Ger-
1
2
3
L
FIG. 1. SDS-polyacrylamide gel electrophoresis (15%) of the cytochrome b&complex after hydroxyapatite chromatography. Marker proteins: Phosphorylase (94 kDa); bovine serum albumin (67 kDa); ovalbumin (43 kDa); carbonic anhydrase (30 kDa); soybean trypsin inhibitor (20.1 kDa); and a-lactalbumin (14.4 kDa).
SPINACH
CHLOROPLAST
CYTOCHROME
f-RIESKE
PROTEIN
SUBCOMPLEX
411
Mol;:i”d~~
a
(KDa)
12
b
3
12
C
3
123456
FIG. 2. SDS-polyacrylamide gel electrophoresis (15%). (a) Cytochromef-Rieske protein complex as obtained from cytochrome b,Jcomplex by partial dissociation with 4 M urea and hydroxyapatite chromatography. Lane 1, marker proteins; lanes 2 and 3, cytochromef-Rieske protein subcomplex. (b) Cytochrome f-Rieske protein subcomplex as obtained by extraction of the cytochrome b6f complex with 50% isopropanol containing 0.1% TFA in the absence of reducing agents. Lane 1, cytochrome bsf complex prepared according to Doyle and Yu (37); lane 2, cytochrome b6f complex purified according to Hurt and Hauska (3); lane 3, cytochrome f-Rieske protein subcomplex. (c) Rieske iron-sulfur protein obtained by solvent extraction of a cytochrome b6f complex which had been reduced with increasing concentrations of mercaptoethanol and aminoethylated. The purified complex was dialyzed, lyophilized, and dissolved in 0.5 M Tris-HCI buffer, pH 8.5, containing 4 M urea, 1% sodium cholate, and 5 mM EDTA, as indicated under Materials and Methods. The final concentration of the complex was equivalent to approximately 10 nmol Cytf/ml. Aliquots of 2 ml were reduced with mercaptoethanol and extracted with 50% isopropanol containing 0.1% TFA. Sequential increased enrichment of the Rieske protein as a function of the reductant concentration was detected by SDS-polyacrylamide gel electrophoresis. Lanes l-6: 0.01, 0.05, 0.1, 0.2, 0.3, and 0.4 mM mercaptoethanol, respectively.
gued that these two polypeptides are perhaps separated rather accidentally because of their similar physical properties rather than subcomplex formation, dissociation of the cytochrome b6f complex by 0.1% SDS and separation by molecular mass using gel filtration in the presence of detergent was attempted. Figure 3a shows that the assumed cytochrome f-Rieske protein subcomplex possessing the highest molecular mass elutes at the front near the void volume, is followed by a second putative binary subcomplex consisting of cytochrome b6 and the 17-kDa protein, and finally by unbound cytochrome b6. Effect of Mercaptoethanol Formation
on Subcomplex
Since the cytochrome b,f complex participates in plants in the electron flow of both photosystems, the effect of reducing
agents like mercaptoethanol on the subcomplex formation appeared interesting. Cytochrome b6f complexes were treated with various concentrations of mercaptoethanol and reacted with ethyleneimine in order to stabilize the reduced state by chemical modification. Samples of thus modified complexes were again extracted with 50% isopropanol (containing 0.1% TFA) and analyzed by SDS-polyacrylamide gel electrophoresis (Fig. 2~). The presence of mercaptoethanol leads to a selective extraction of almost pure Rieske iron-sulfur protein with only traces of cytochrome f present. Figure 2c shows sequential increased enrichment of the FeS protein at increasing concentrations of the reductant. An analogous experiment involving gel filtration of the cytochrome b&complex in 0.1% SDS in the presence of mercaptoethanol was performed (Fig. 3b). In contrast
412
EL-DEMERDASH,
SALNIKOW,
AND
VATER
Cyt f------...----CYt b6 --___---__
Rieske
-FeS-
Direction
of
elution.-w
FIG. 3. SDS-polyacrylamide eleetrophoresis (15%) of fractions obtained from gel filtration of the cytochrome &fcomplex on AcA-54 in the presence of 0.1% SDS (a) under nonreducing conditions, or (b) in the presence of 0.5% mercaptoethanol.
to the subcomplex pattern observed under nonreducing conditions the four polypeptides now eluted in order of their molecular masses. These results indicate that the binary complex obtained in the absence of mercaptoethanol represents a native building block of the cytochrome b,f architecture stabilized by molecular forces which are sensitive toward reducing agents. DISCUSSION
In this study the cytochrome b,f complex from spinach chloroplast was prepared according to the method of Hurt and Hauska (3), however, the time-consuming final ultracentrifugation step was replaced by hydroxyapatite chromatography as recommended by Doyle and Yu (37) yielding considerable gains in time and sample capacity. Recently, it was shown that by using this method active cytochrome b&complexes could be prepared on a large scale in at least 90% pure form
(38). The polypeptide composition of our preparation is in close agreement with other published data (3,37-39). The dissociation studies in the cytochrome b6f system using mild denaturants such as 4 M urea or 0.1% SDS as described in this report point to a subarchitecture of the entire complex consisting of the binary building blocks cytochrome fRieske protein and cytochrome bs-1’7-kDa protein (= subunit 4 of the cytochrome bsf complex). A functional linkage of cytochrome f and the Rieske iron-sulfur protein has been deduced by several authors (5,26-28); however, isolation of a binary molecular complex proving the molecular proximity of both components has to our knowledge been demonstrated here for the first time. Thus, the cytochrome f-Rieske protein complex represents another link in the chain of adjacent electron carriers for which interaction has been demonstrated such as cytochrome f/plastocyanin (31-34) or ferredoxin/ferredoxin-NADP+ reductase pairs (40-43). In
SPINACH
CHLOROPLAST
CYTOCHROME
mitochondria, the Rieske iron-sulfur protein is closely associated with cytochrome cl which is functionally the biological equivalent of cytochrome f in this organelle. It has been shown recently that both components interact tightly in the complex III of beef heart mitochondria (44) thus constituting an integral part of the electron pathway from ubiquinol to cytochrome c. In view of the evolutionary resemblance of the chloroplasts and mitochondrial organelles-specifically the electron carrier systems involved in energy transduction-it is not surprising that the molecular environment elucidated for the mitochondrial Rieske ironsulfur protein is mirrored to some extent in the chloroplast system. A second observation evident from the gel filtration experiment is a close association of cytochrome b6 with the l’i-kDa protein. Again, bearing in mind the structural similarities between the mitochondrial complex III and the chloroplast b6f complex, this result is not unexpected, since the l’l-kDa protein shows considerable sequence homology to the C-terminal end of mitochondrial cytochrome b, thus pointing to a common ancestor gene early in the evolution (45). The forces involved in the interaction of cytochrome f and the Rieske iron-sulfur protein are at present not clear. Since complete dissociation of the cytochrome b&complex in the presence of detergents requires reducing conditions as, for example, mercaptoethanol, apparently not only hydrophobic interactions are responsible for the stability of the subcomplex, but disulfide bridges seem to play an important role as well. However, taking into account available sequence data and present concepts of thylakoid membrane organization a direct disulfide linkage between cytochrome f and the Rieske iron-sulfur protein appears difficult to formulate: the major N-terminal part of cytochrome f which is exposed to the lumen of the thylakoid membrane possesses only two cysteine residues (46) which are presumed to serve as heme attachment sites thus eliminating possible additional thiol linkages. The Rieske protein of plant chloroplasts,
f-RIESKE
PROTEIN
SUBCOMPLEX
413
on the other hand, as analyzed by sequencing of the 96 N-terminal residues and isolation of additional peptide fragments possesses five cysteinyl residues (47): four cysteins are found in conserved sequences which match with those determined in the C-termini of Rieske proteins from fungal and bacterial cytochrome bcl complexes (48,49) and which are thought to represent the (2Fe-2s) cluster binding sites; an additional cysteinyl peptide has been identified which apparently does not participate in iron-sulfur core binding (48). The requirement for four cysteines in iron binding, however, is not yet certain in the chloroplast system; recent studies on the respiratory chain of thermophilic bacteria point to nitrogen atoms as additional coordination sites for iron (50, 51) thus limiting the number of essential cysteinyl side chain functions for cluster formation to two. Consequently, the chloroplast Rieske iron-sulfur protein possesses one to possibly three potential thiol functions qualifying for disulfide bridge formation. The observation of a cytochrome fRieske protein subcomplex necessitates a discussion of the structural organization of the cytochrome b6f complex in the thylakoid membrane, although solid data are scarce here and the models are somewhat controversial compared to the analogous mitochondrial cytochrome beI systems. There is ample evidence that in the cytochrome beI complex the Rieske iron-sulfur protein and cytochrome cl traverse the membrane displaying their functional groups in close proximity (18, 44); the amino acid sequences of the Rieske proteins from the cytochrome bq complex of Neurospora and Rhodopseudomonas sphaeroides (48, 49) show a hydrophobic segment at the N-terminal half long enough to span the membrane barrier anchoring the protein in it; the iron-sulfur cluster linked to the more hydrophilic Cterminal portion would be exposed into the intermembrane space on the inner membrane surface. Sequence data of the chloroplast Rieske protein (47) point to an analogous membrane orientation involving N-terminal anchoring in the membrane via a hydrophobic membrane span-
414
EL-DEMERDASH,
SALNIKOW,
ning segment. The iron-sulfur cluster as the prosthetic group of the catalytic domain would be located at the C-terminal part and, consequently, exposed toward the thylakoid lumen, thus placing the Nterminus toward the stromal surface. This orientation would permit close contact of the iron-sulfur cluster of the Rieske protein with the similarly oriented heme function of cytochrome $ In support of this orientation antibodies against the Rieske protein have been shown to not inhibit electron transport from water or plastoquinol to photosystem I in chloroplasts (24). In contrast to this model, studies using limited proteolysis and specific antibodies favor an opposite orientation of the Rieske protein with the putative ironsulfur cluster carrying the C-terminal fragment close to or on the stromal side of the thylakoid membrane surface (52). This model, however, places severe spatial constraints on the functional heme-iron-sulfur interaction required for electron flow. The dissociating effect of reducing agents such as mercaptoethanol could perhaps be a function of the iron-sulfur core of the Rieske protein itself. Under dissociating conditions such as with 0.1% SDS the iron-sulfur cluster is fully retained as has been demonstrated by Nishikimi et al. (53) permitting subsequent reactions with redox reagents. Furthermore, it has been shown that the Rieske protein is more readily extracted from the mitochondrial complex III under reducing conditions (54) pointing again to an involvement of thiol groups in binding. In the cytochrome bq system a tenacious association of the Rieske protein with cytochrome cl has been observed; final separation was achieved by passage of the partially purified system through a mercurial resin column which retained the cytochrome component, however, not the Rieske protein (55), indicating that the latter does not possess free SH groups on its surface when complexed to cytochrome cl. On the other hand, cleavage of the total cytochrome reductase complex in Triton X-100 by higher salt concentrations in the presence of ascorbate and dithioerythritol leads to facile dissociation yielding the
AND
VATER
iron-sulfur subunit and complexes lacking redox centers as separable components (56). Considering the structural and functional similarities of the cytochrome bq and cytochrome b6f systems, these observations appear as additional support for the experimental data described in this report. In summary, the available information on the structure of the cytochrome b6f complex and the stability of the cytochrome f-Rieske protein subcomplex can perhaps be explained by an intramolecular disulfide bridge of the Rieske protein itself stabilizing a conformation required for tight interaction of the heme and iron-sulfur chromophores; this would be in accordance with the structural and three-dimensional features of current models. In view of a lack of additional supportive evidence, however, this hypothesis remains largely speculative at present. REFERENCES 1. TRUMPOWER, B. L., AND KATKI, A. G. (1979) in Membrane Proteins in Energy Transduction (Capaldi, R. A., Ed.), pp. 89-200, Dekker, Basel/New York. 2. NELSON, N., AND NEUMANN, J. (1972) J. Biol. Chem. 247,1917-1924. 3. HURT, E., AND HAUSKA, G. (1981) Eur. J. Biothem. 117,591-599. 4. CROFTS, A. R., MEINHARDT, S. W., AND BOWYER, J. R. (1981) in Function of Quinones in Energy Conversing Systems (Trumpower, B. L., Ed.), Academic Press, New York. 5. HAUSKA, G., HURT, E., GABELLINI, N., AND LOCKAU, W. (1983) Biochim Biophys. Acta 726, 97-133. 6. BARBER, J. (1984) Trends B&hem Sci. 9,209-210. 7. HAEHNEL, W. (1984) Annu. Rev. Plant. Physiol 35,659-693. 8. HURT, E., AND HAUSKA, G. (1982) Biochim. Biophys. Acta 682,466-473. 9. MORSCHEL, E., AND STAEHELIN, L. A. (1983) J. Cell Biol. 97, 301-310. 10. WILLEY, D. L., AUFFRET, A. D., AND GRAY, J. C. (1984) Cell 36,555-562. 11. MANSFIELD, R. W., AND BENDALL, D. S. (1984) B&him Biophys. Acta 766,62-69. 12. MANSFIELD, R. W., AND ANDERSON, J. M. (1985) Biochim. Biophys. Acta 809.435-444. 13. WILLEY, D. L., AND GARY, J. C. (1984) in Advances in Photosynthesis Research (Sybesma,
SPINACH
14.
15. 16. 17.
18. 19.
20.
21.
22. 23. 24. 25.
26. 27.
28.
29. 30. 31.
CHLOROPLAST
CYTOCHROME
C., Ed.), Vol. IV, pp. 567-570, Nijhoff, Junk, The Hague. ALAM, J., AND KROGMANN, I. W. (1984) in Advances in Photosynthesis Research (Sybesma, C., Ed.), Vol. I, pp. 521-524, Nijhoff, Junk, The Hague. HURT, E. (1983) Dissertation, University of Regensburg. GELLERFORS, P., AND NELSON, B. D. (1978) FEBS Lett. 92.36-40. BELL, R. L., SWEETLAND, J., LUDWIG, B., AND CAPALDI, R. A. (1979) Proc. N&l. Acad. Sci. USA 76,741-745. LI, Y., DE VRIES, S., LEONARD, K., AND WEISS, H. (1981) FEBS Lett. 135,277-280. LEONARD, K., HOVM~LLER, S., KARLSSON, B., WINGFIELD, P., LI, Y., PERKINS, S., AND WEISS, H. (1981) in Vectorial Research in Electron and Ion Transport in Mitochondria and Bacteria (Palmieri, F., Quagliarello, E., Siliprandi, N., and Slater, E. C., Eds.), pp. 129-137, Elsevier/North-Holland, Amsterdam. SHNIEDER, D. L., AND RACKER, E. (1973) in Oxidases and Related Redox Systems (King, T. E., Mason, H. S., and Marrison, M., Eds.), pp. 799-821, Wiley, New York. BEATTIE, D. S., SIDHU, A., AND CLEJAN, L. (1982) in Proceedings, 2nd European Bioengineering Conference Short report, pp. 203-204, University Claude Beranard, Lyon. SIDHU, A., AND BEATTIE, D. S. (1982) J. Biol. Chem. 257,7879-7886. Biophys. BOUGES-BOCQUET, B. (1980) Biochim. Acta 594, 85-103. HURT, E., HAUSKA, G., AND MALKIN, R. (1981) FEBS Lett. 134, l-5. OETTMEIER, W., MASSON, K., SOLL, H.-J., HURT, E., AND HAUSKA, G. (1982) FEBS Lett. 144, 313-317. VELTHUYS, B. R. (1980) Annu. Rev. Plant Physiol. 31,545-567. MALKIN, R. (1982) in Electron Transport and Phosphorylation (Barber, J., Ed.), Vol. 4, pp. 2-47, Elsevier Biochemical, Amsterdam/New York. CRAMER, W. A., AND CROFTS, A. R. (1982) in Photosynthesis (Govindjee, Ed.), Vol. I, pp. 387-467, Academic Press, New York. LAM, E., AND MALKIN, R. (1982) FEBS Lett. 144, 190-193. HAEHNEL, W., BERZBORN, R. J., AND ANDERSSON, B. (1981) B&him. Biophys. Acta 637,389-399. WOOD, P. M. (1974) Biochim. Biophys. Acta 357, 370-379.
f-RIESKE
PROTEIN
SUBCOMPLEX
415
32. NIWA, S., ISCHIKAWA, H., NIKAI, S., AND TABAKE, T. (1980) J. Biol. Chem. 88,1177-1183. 33. TANAKE, K., TAKAHASHI, M., AND ASADA, K. (1981) Plant Cell PhysioL 22,33-39. 34. DAVIS, D. J., AND HOUGH, K. (1983) Biochem. Biophys. Res. Cammun. 116,1000-1006. 35. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. Biol. Chem. 193, 265-275. 36. LAEMMLI, U. K. (1979) Nature (London) 227, 680-685. 37. DOYLE, M. F., AND Yu, C.-A. (1985) Biochem. Biophys. Res. Commun. 131,700-706. 38. BLACK, M. T., WIDGER, W. R., AND CRAMER, W. A. (1987) Arch. Biochem. Biophys. 252,655-661. 39. CLARK, R. D., AND HIND, G. (1983) J. Biol. Chem. 258,10348-10354. 40. SHIN, M., AND SAN PIETRO, A. (1968) Biochem. Biophys. Res. Cwmmun. 33,38-42. 41. FOUST, G. P., MAYHEW, S. G., AND MASSEY, V. (1969) J. Biol. Chem. 244,964-970. 42. BAGER, P. (1971) Planta 99,319-338. 43. SHIN, M. (1973) Biochim Biophys. Acta 292, 13-19. 44. SHIMOMURA, Y., NISHIKIMI, M., AND OZAWA, T. (1985) J. BioZ. Chem. 260,15075-15080. 45. WIDGER, W. R., CRAMER, W. A., HERRMANN, R. G., AND TREBST, A. (1984) Proc. Natl. Acad Sci. USA 81,674-678. 46. ALT, J., AND HERRMANN, R. G. (1984) Curr. Genet. 8,551-557. 47. PFEFFERKORN, B., AND MEYER, H. E. (1986) FEBS Lett. 206,233-237. 48. HARNISCH, K., WEISS, H., AND SEBALD, W. (1985) Eur. J. B&hem. 149,95-99. 49. GABELLINI, N., AND SEBALD, W. (1986) Eur. J. Biochem. 154,569-579. 50. FEE, J. A., FINDLING, K. L., YOSHIDA, T., HILLE, R., TARR, G. E., HEARCHEN, D. O., DUNHAM, W. R., DAY, E. P., KENT, T. A., AND M~NCK, E. (1984) J. Biol. Chem. 259,124-133. 51. KUILA, D., AND FEE, J. A. (1986) J. Biol. Chem. 261,2768-2771. 52. MANSFIELD, R. W., AND ANDERSON, J. M. (1985) Biochim. Biophys. Acta 809,435-444. 53. NISHIKIMI, M., SHIMOMURA, Y., YAMADA, H., AND OZAWA, T. (1984) Biochem. Bioph,ys. Res. Cammm. 120,237-241. 54. RIESKE, J. S. (1976) Biochim. Biophys. Acta 456, 195-247. 55. TRUMPOWER, B. L., AND EDWARDS, C. A. (1979) J. Biol. Chem. 254,8697-8706. 56. HOVM~LLER, S., LEONARD, K., AND WEISS, H. (1981) FEBS Lett. 123,118-122.