Current address: D. Baniulis, Lithuanian Institute of Horticulture, Babtai, Kaunas reg., Lithuania.
Structure–Function of the Cytochrome b6f Complex†
Article first published online: 7 OCT 2008
© 2008 The Authors. Journal Compilation. The American Society of Photobiology
Photochemistry and Photobiology
Volume 84, Issue 6, pages 1349–1358, November/December 2008
How to Cite
Baniulis, D., Yamashita, E., Zhang, H., Hasan, S. S. and Cramer, W. A. (2008), Structure–Function of the Cytochrome b6f Complex. Photochemistry and Photobiology, 84: 1349–1358. doi: 10.1111/j.1751-1097.2008.00444.x
This invited paper is part of the Symposium-in-Print: Photosynthesis.
- Issue published online: 3 NOV 2008
- Article first published online: 7 OCT 2008
- Received 20 May 2008, accepted 30 July 2008
- Top of page
- The Problem of Proteolysis
The structure and function of the cytochrome b6 f complex is considered in the context of recent crystal structures of the complex as an eight subunit, 220 kDa symmetric dimeric complex obtained from the thermophilic cyanobacterium, Mastigocladus laminosus, and the green alga, Chlamydomonas reinhardtii. A major problem confronted in crystallization of the cyanobacterial complex, proteolysis of three of the subunits, is discussed along with initial efforts to identify the protease. The evolution of these cytochrome complexes is illustrated by conservation of the hydrophobic heme-binding transmembrane domain of the cyt b polypeptide between b6 f and bc1 complexes, and the rubredoxin-like membrane proximal domain of the Rieske [2Fe-2S] protein. Pathways of coupled electron and proton transfer are discussed in the framework of a modified Q cycle, in which the heme cn, not found in the bc1 complex, but electronically tightly coupled to the heme bn of the b6 f complex, is included. Crystal structures of the cyanobacterial complex with the quinone analogue inhibitors, NQNO or tridecyl-stigmatellin, show the latter to be ligands of heme cn, implicating heme cn as an n-side plastoquinone reductase. Existing questions include (a) the details of the shuttle of: (i) the [2Fe-2S] protein between the membrane-bound PQH2 electron/H+ donor and the cytochrome f acceptor to complete the p-side electron transfer circuit; (ii) PQ/PQH2 between n- and p-sides of the complex across the intermonomer quinone exchange cavity, through the narrow portal connecting the cavity with the p-side [2Fe-2S] niche; (b) the role of the n-side of the b6 f complex and heme cn in regulation of the relative rates of noncyclic and cyclic electron transfer. The likely presence of cyclic electron transport in the b6 f complex, and of heme cn in the firmicute bc complex suggests the concept that hemes bn-cn define a branch point in bc complexes that can support electron transport pathways that differ in detail from the Q cycle supported by the bc1 complex.
- Top of page
- The Problem of Proteolysis
Electron transport pathways, information from crystal structures, the 3.0 Å barrier
The cytochrome b6 f complex is one of the three hetero-oligomeric integral membrane protein complexes responsible for electron transport and energy transduction in oxygenic photosynthetic membranes. The b6 f complex occupies an electrochemically central position in the noncyclic or “linear” electron-transport chain (1,2). The plastoquinol-plastocyanin/cyt c6 oxidoreductase activity of the complex provides an electronic connection between the two photosynthetic reaction center complexes, photosystems I and II (Fig. 1), in which electron transfer through the complex is coupled to proton transfer from the electrochemically negative (n) to the positive (p) side of the complex. The major tasks of the studies on structure–function of this complex are to understand the biochemical mechanisms of the H+/e− coupling on the n- and p-sides of the complex. These mechanisms probably differ to some extent from those proposed for the partly homologous bc1 complex in the mitochondrial respiratory chain (e.g.3–10) and the bc1 complex in the photosynthetic bacterium, Rhodobacter sphaeroides (11) because (a) the b6 f complex participates in the PSI-linked cyclic electron transport chain (Fig. 1), in which ferredoxin (12–16) and ferredoxin: NADP+ reductase (FNR) (17–20) can function as electron transfer intermediates between PSI and the b6 f complex, as does a small very basic peptide, “PGR5” (21), presumably for reasons of structure. The inference that the b6 f complex, and not another membrane protein (22) provides the function of a ferredoxin:plastoquinone reductase in cyclic electron transport is supported by the presence of bound FNR at a significant stoichiometry in purified spinach b6 f complex (19,20), the requirement of both ferredoxin and FNR for NADPH reduction of heme b in the complex (16), and by the absence of evidence for any other ubiquitous membrane protein complex to serve this function. However, a rigorous proof for competent kinetics of electron transfer from PSI to the b6 f complex is missing. (b) The other redox prosthetic group not found in the bc1 complex is the heme cn (23–25), which was found in the crystal structures from both cyanobacterial and green algal sources, to be covalently bound to the cyt b6 polypeptide on the n-side of the complex (26,27). The properties of this unique heme are discussed below at greater length.
The development of concepts concerning the mechanism of electron and proton transfer in the b6 f complex has been catalyzed by crystal structures of the b6 f complex from the filamentous thermophilic cyanobacterium, Mastigocladus laminosus (16,26,28) and the green alga, Chlamydomonas reinhardtii (27). A wire diagram of the 3.0 Å structure of the native dimeric b6 f complex from M. laminosus, crystallized in the presence of cadmium (Cd2+) ion, added for additional stabilization, is shown (Fig. 2) (16).
The structures from M. laminosus and C. reinhardtii, while supporting major new insights into structure–function of the b6 f complex, have a limitation in the information that they have been able to provide because the resolution of the native structure, and of structures obtained in the presence of quinone analogue inhibitors, presently does not extend beyond the resolution of 3.0 Å (16) and 3.1 Å (27), respectively, obtained for native and inhibitor complexes from M. laminosus, and for the complex from C. reinhardtii with the QAI, tridecyl-stigmatellin. This resolution is not sufficient to allow accurate determination of the orientation of amino acid side chains. Thus, future progress depends on the ability to “transcend this 3.0 Å barrier.”
The Problem of Proteolysis
- Top of page
- The Problem of Proteolysis
In the case of the cyanobacterial complex, a major problem in the crystal structure studies has been proteolysis and resulting monomerization of the complex (29), which degrades the complex from its active dimeric form. This proteolysis problem has precluded the use of unicellular cyanobacteria such as Synechocystis PCC 6803 and Synechococcus PCC 7102, for which the genome sequence is known and mutagenesis is possible, as sources for crystal structures of the b6 f complex because monomerized inactive complex will not crystallize. Mutagenesis experiments on the b6 f complex have been carried out in the unicellular cyanobacteria, Synechococcus sp. 7002 (30,31) or Synechocystis sp PCC 6803 (32). These cyanobacteria are readily grown, and in principle should be an effective protein expression system for obtaining stable photosynthetic membrane proteins for structural studies. However, protein structure analysis and accurate determination of specific activity requires a high degree of protein purity in the cytochrome complex extracted, isolated and purified from membranes and in detergent solutions. Thus far, it has not been possible to purify in significant yield an active dimeric b6 f complex from the unicellular cyanobacteria. Cytochrome b6f complex extracted with detergent from these sources is always monomerized, as shown by the migration profile in a sucrose density gradient (Fig. 3), and thereby inactivated.
Proteolysis of the b6 f complex was also observed in the filamentous M. laminosus, but it occurred slowly enough that crystals could be obtained when the crystallization process was accelerated through addition of lipid to the purified delipidated complex (33). M. laminosus is not transformable at present, and therefore cannot support site-directed mutagenesis studies of the cyt b6f complex in cyanobacteria.
The reasons for the monomerization of the cyt b6 f dimer upon extraction from the membrane are not known. Proteolytic degradation of three subunits in the purified complex from M. laminosus has been observed, suggesting that preparations from other cyanobacteria might contain a damaging proteolytic activity. In addition, the instability may be enhanced by intrinsic instability of the protein complex in detergent micelles because of incompatible detergent or over-purification resulting in depletion of lipids required for stabilization of protein structure.
The amphipathic feature of detergents is the basis for their efficacy as solubilizing agents of integral membrane proteins (34,35). Detergents with small head groups and short alkyl chains are more suitable for solubilization. Those with larger head groups and longer alkyl chains are better for stabilizing membrane proteins. Detergent compatibility is an important factor affecting stability of the b6f complex during extraction and subsequent purification. The method originally used to isolate cyt b6f complex from spinach thylakoid membranes (36) was modified to isolate the complex from the filamentous cyanobacterium M. laminosus (33,37). However, the method has thus far not been applicable for isolation of stable b6f complex from unicellular cyanobacterial strains. Use of a procedure for histidine-tagged complex purification from Synechococcus sp PCC 7002 that employed the “mild”“Hecameg” or amphiphilic detergents such as undecyl- or dodecyl-maltopyranosides, did not improve the stability of the complex, and detergent extraction resulted in loss of protein activity (J. Yan and H. Zhang, unpublished observations).
Extensive purification of an integral membrane protein complex may lead to lipid depletion and resulting instability. This problem was observed in the crystallization of the b6f complex (33). In the early stages of crystallization trials, it was found that an extensive purification of the complex led to delipidation. The purified b6 f complex in the final stage was found to contain <0.5 molecules of mono-galactosyl-diacyl glycerol of bound lipid per monomer. Lipid depletion caused instability of the b6 f complex. No well diffracting crystals ever formed from these preparations. Addition of DOPC or DOPG lipid to the complex immediately after the last step in the purification at a stoichiometry of 10:1, lipid:cyt f, resulted in formation of hexagonal crystals. The resolved cyt b6 f structure demonstrated a structural role of lipid in the complex (26,27). Three lipid molecules, a sulfo-lipid (27) and two DOPC molecules presumably arising from the lipid added before crystallization that probably displaces endogenous lipid, can be resolved in the intermonomer quinone exchange cavity (Fig. 4). The structural importance of lipids in energy transducing cytochrome complexes has been well documented for the mitochondrial cyt bc1 (38,39) and cytochrome oxidase (40) complexes.
However, it is likely that presence of the lipids in the structure is not the only factor required to ensure dimer stability, since the dimeric b6 f complex is retained prior to lipid augmentation during the purification procedure from M. laminosus.
The presence of protease even in trace amounts during the protein preparation could be destructive to the integrity of the complex, and a cause of instability. A mixture of protease inhibitors is commonly employed in all solutions used for purification. However, inhibitor activity may not be effective against certain types of proteolytic enzymes or over extended time intervals. Examination by SDS-PAGE of the M. laminosus b6f complex, which was kept in different detergents at room temperature for 7–14 days, showed proteolysis (29). This was most obvious for the [2Fe–2S]-binding subunit, from which a 48-amino acid C-terminal fragment was cleaved. The cyt b6 f and subunit IV polypeptides were also clipped at their exposed termini. Stability tests carried out on the complex at room temperature showed proteolysis in about 1 week, even in the presence of protease inhibitors. Protease inhibitors at moderate concentration could retard but not inhibit the proteolysis.
The detailed mechanism of proteolytic degradation of purified b6f complex has not been characterized. However, it is assumed that the proteolytic activity involves either a non-specific mechanism that is triggered by activation, release of proteases after cell breakage, or specific mechanisms that are related to peptide processing peptidases of biogenesis pathways. It has been demonstrated that the serine-type protease ClpP controls a proteolytic processing of fully and partially assembled cyt b6f during biogenesis in C. reinhardtii (41). A multigene clp family has been characterized in cyanobacteria (42). Clp proteins in eubacteria are induced by external environmental stress. ClpX and ClpP are typically induced by high temperatures (43,44) and other stresses such as high salt, oxidation and glucose deprivation.
Recently, we obtained successful isolation and crystals of an active dimeric cytochrome b6f complex from the filamentous cyanobacterium Nostoc (Anabaena) sp. PCC 7120 (D. Baniulis et al., unpublished data) whose distinguishing property compared with M. laminosus is that it is transformable. Proteolytic enzyme composition analysis of the Nostoc sp. PCC 7120 genome, using the MEROPS peptidase database, provided additional clues to support the inference that proteolytic cleavage is responsible for destabilization of the cyt b6f dimer. The peptidase composition of the Nostoc 7120 strain was compared to that of selected cyanobacterial strains (Thermosynechococcus elongatus, Anabaena variabilis, Synechococcus elongatus, and Synechocystis sp. PCC 6803), which we and/or others had previously used for purification of the b6 f complex. In addition, several other more commonly studied strains representing different cyanobacterial subclasses and other types of bacteria with comparable genome size were included in the analysis. Four unique peptidases were found in analyzed cyanobacteria strains compared with Nostoc sp. PCC 7120 (Table 1). If it is assumed that the same peptidase(s) is (are) responsible for the cyt b6 f cleavage in all the cyanobacteria, then it must be among the peptidases that are unique to T. elongatus from which the b6 f complex was found to be proteolyzed These include (N.B. number in parentheses is MEROPS peptidase database identifier), glucosamine-fructose-6-phosphate aminotransferase (C44.971), and the microcin-processing peptidase 2 (U62.002), which are found in all four cyanobacterial strains that yield monomerized complex. Activation of one or both of the proteases could be responsible for the cleavage of subunits of the b6f complex during detergent extraction and purification.
|Strain||Genome size, Mbp||Total, known or putative peptidases||Unique peptidases compared to Nostoc 7120|
|Nostoc sp PCC 7120 (IV)a||7.21||125||–|
|Anabaena variabilis (IV)b||7.07||157||20|
|Trichodesmium erythraeum (III)||7.8||157||28|
|Gloeobacter violaceus (I)||4.66||118||25|
|Prochlorococcus marinus (I)||1.6||386||94|
|Synechococcus elongatus (I)b||2.7||100||20|
|Synechocystis sp. PCC 6803 (I)b||3.95||78||11|
|Thermosynechococcus elongatus (I)b||2.59||52||4|
|Escherichia coli K12||4.6||226||142|
- Top of page
- The Problem of Proteolysis
Unique prosthetic groups and their properties
Four bound prosthetic groups are common to the b6 f and bc1 complexes (Fig. 5): two noncovalently bound b-type hemes (hemes bp and bn on the p- and n-sides of the complex), a high potential [2Fe-2S] cluster in the Rieske protein, and the covalently bound c-type heme of cytochromes c1 and f. Hemes bp and bn, in the core of the complex that is structurally conserved in evolution between bc1 and b6 f complexes (45), bridge the second and fourth transmembrane helices of the cyt b polypeptide. Three prosthetic groups in the b6 f complex, the heme cn (Fig. 6) discussed above, one chlorophyll a and one β-carotene molecule per b6 f monomer, are uniquely found in the b6 f complex, but not in the bc1 complex from either the respiratory chain (Fig. 5B) or photosynthetic bacteria. The heme cn is unique among protein bound hemes in not having any amino acid side chain as an axial ligand. The crystal structure shows that it is positioned very close to the heme bn, as the distance between the cn heme Fe and a propionate of heme bn, bridged by an H2O or OH−, is only 3.5–4.0 Å. The structural proximity between hemes bn and cn is reflected in strong electronic coupling between the two hemes as shown by orthogonal and parallel mode EPR spectra of the isolated complex with g values as high as 12 (46). Similar results in the orthogonal mode were also obtained with the b6 f complex from C. reinhardtii (47). (N.B. heme cn is called heme ci in the studies on the b6 f complex from C. reinhardtii. The choice of the notation for this heme has been discussed ).
The molecular sizes of the eight subunits of the b6 f complex from M. laminosus, determined by electrospray mass spectrometry (48), pI values of these subunits, and the midpoint redox potential of the hemes and [2Fe-2S] cluster of cytochromes f, b, and the ISP, are summarized (Table 2). The three largest subunits, petA, B and D contain bound redox prosthetic groups. The four small (MW ≅ 3–4 kDa) subunits are bound as four “hydrophobic sticks” on the periphery of each monomer whose insertion into the complex and the membrane obeys the cis-positive rule (49,50) with positively charged residues in these inserted subunits only on the n-side of the complex. These small subunits and their arrangement have no structural analogy in the bc1 complex. The edge–edge-distances between the prosthetic groups in the complex from M. laminosus are summarized (Table 3).
|Subunit||Measured mass (Da) (48)||pI||Em,7 (mV)|
|Cyt b6||24712||9.0||−50 (bn), −50 to 150 (bp), ∼ +50 to +100 (heme cn)*|
|Prosthetic group||Distance (Å)|
|Fe to Fe||Edge to edge|
|Heme bp—Heme bn||20||8|
|Heme bp—Heme bp||22||11|
|Heme bn—Heme bn||35||29|
|Heme bn—Heme cn||9.6||4 (bn Fe to cn propionate)|
|Chl a—Heme bp||_||12|
|Chl a—Heme bn||_||5.5|
Among the major questions that concern the function of the b6 f complex are: (a) the function of the three unique prosthetic groups mentioned above; (b) whether the complex functions in noncyclic/linear flow through a “Q cycle” that is obligatory; (c) whether it is an obligatory intermediate in ferredoxin- and photosystem I–mediated cyclic electron transport and if so, how cyclic and noncyclic electron transport are regulated (“time-shared”). The purpose of the reactions summarized in Table 4 is to raise the question of differences in the quinone-based electron transfer pathways in the b6 fvs. the bc1 complex. The literature on this subject contains the viewpoint that a Q cycle similar to that described for the bc1 complex is obligatory in the b6 f complex (9), or that it may be facultative and dependent upon the magnitude of the proton electrochemical potential gradient (51).
|A1. p-side quinol oxidation*|
|A2. high potential chain|
|A3. transmembrane p- to n-side electron transfer|
|A4. n-side 2 x 1 electron reduction of PQ (as in bc1 complex)|
|B. n-side 2 electron reduction of bound PQ using 2 electrons from p-side; intermonomer PQH2|
|(i) supply two electrons cooperatively to PQ after each heme is reduced by two lumenal p-side turnovers of|
|C. n-side 2 electron reduction of PQ by 1 electron from p-side, 1 from Fd (n-side)|
Direct application of the Q cycle formalism is shown (Fig. 7, Table 4 A.1–4) (4,52,53). An altered set of reactions that would include the heme cn that is absent in the bc1 complex and the ferredoxin-linked cyclic pathway is presented (Table 4C). A one-electron transfer from PQH2 to the [2Fe-2S] cluster can generate an anionic semiquinone radical (PQ•−) of more negative redox potential that reduces the heme bp, for which the Em ≅ −50 mV in the b6 f complex (54,55), and thereby initiates electron transfer across the complex through a lower potential chain, heme bp (r) + heme bn (o) bp (o) + bn (r) [Table 4A.1]. One problem with this model is that the PQ•− species has not been routinely detected by EPR, although some supportive evidence has been presented (56). The absence of extensive EPR data raises the question of the stability and population density of the PQ•− at the p-side near the [2Fe-2S] cluster. Instability of semiquinone, whether it is quenched by short-circuit reactions involving heme bp, and whether the mechanism may bypass the one-electron nature associated with the semiquinone in favor of a two-electron concerted reaction, is discussed (7,57). The existence of a significant population of semiquinone at the p-side is implied by generation of superoxide in a “bypass” reaction (58), at least at a rate ∼1% of normal electron transfer.
One perspective that seems to be missing thus far from the literature discussion of the p-side events in Q/QH2 reactions is that of insertion/extrusion of Q/QH2 to/from the [2Fe-2S] cluster through the p-side portal. The protrusion of tridecyl-stigmatellin and the chlorophyll phytyl tail from the portal, derived from the cyanobacterial crystal structures (16) is shown (Fig. 8) (1,31). The forces that guide the reversible movement and the time of residence of Q/QH2 in the portal are parameters relevant to the nature of their p-side interface redox reactions, but which have not yet been analyzed.
In the Q cycle model, two turnovers of the quinol, PQH2, on the lumenal p-side result in the transfer of two electrons through heme bp to a stromal n-side quinone bound near heme bn. The position of the p- and n-side redox components essential for the Q cycle in the bc1 complex (52,59) has been supported not only by extensive spectrophotometric and kinetic data (3,5–7,10,60,61), but also by crystal structure data that confirm the predicted positions of the stromal n-side ubiquinone relative to heme bn, and the positions of p- (62) and n-side (63) quinone analogue inhibitors. On the p- and n-sides of the bc1 complex, stigmatellin and antimycin A, respectively, displace the redox-active quinone as predicted by the Q-cycle model. In the b6 f complex, tridecyl-stigmatellin binds on the lumenal p-side in an analogous manner (26,27), as well as binding as a heme cn axial ligand on the n-side. In b6 f, the n-side inhibitor, NQNO, binds more weakly than does antimycin in the bc1 complex and at a site distinctly different (16). For this reason, and because the two b-hemes have very similar absorbance spectra in the b6 f complex, the evidence for function of the b6 f complex in a Q cycle (9,64,65) is less complete.
A scheme for coupled electron and proton transport shown in Table 4 (A1–A4) demonstrates the standard Q cycle model applied to the b6 f complex. The stromal n-side reduction of PQ proceeds in two 1-electron steps via a plastosemiquinone. In contrast, Table 4B presents a Q cycle model for stromal n-side PQ reduction modified as a result of the concerted two-electron transfer from the tightly coupled hemes bn/cn (see section Structure–function above). In this model, the PQ binding site in a bn/cn/PQ complex is emphasized, as is release of PQH2 from this site to the intermonomer space of the quinone-exchange cavity. The reaction sequence shown in Table 4C emphasizes the possibility of a contribution to the n-side electron transport reactions of one electron donation from ferredoxin (Fd), participating in PSI-linked cyclic electron transfer (1,16,20) to the stromal bn/cn/PQ complex. One electron reduction of PQ by Fd (reduced), and resultant formation of a more reducing quinone species that could form the reactive oxygen species, the superoxide anion radical, , could be prevented by a two-electron gate mechanism imposed by the coupled hemes bn/cn.
Structurally distinct events in the electron transfer cycle of bc complexes (Table 4); the rate-limiting step
In the first crystal structure of the bc1 complex, it was found that the distance between the [2Fe-2S] cluster bound close to the quinone at the membrane interface and the heme of cytochrome c1 was greater than 30 Å (66). This donor–acceptor distance for electron transfer in the high potential chain is orders of magnitude greater (67,68) than that which would be compatible with the known rate-limiting step (milliseconds) of the respiratory chain, implying that either cyt c1 or the ISP must move to bridge this distance. A second crystal form was subsequently found in which the [2Fe-2S] cluster had undergone a 16–17 Å translation, and the ISP a 60º rotation to bring the cluster close enough to the heme c1 for competent electron transfer (69). The structures of the cyt b6 f complex show the same incompatibility with function of the distance between the membrane-proximal [2Fe-2S] cluster and the cyt f heme. A second crystal form has not yet been found for the b6 f complex. Because of the different structure of cyt f and position of the cyt f heme, it was proposed that required rotation of the ISP associated with the millisecond electron transfer chain would be 25º. The possibility that this large conformational change of the Rieske protein could be associated with the rate-limiting step has been discussed and dismissed partly because as a simple diffusion event, it would span the ∼20 Å distance in a sub-microsecond time (70). This estimate of the transfer time is different if a restrained diffusion, which is more realistic, is considered. The latter consideration is implicit in an analysis of structure barriers to rotation-translation of the ISP (71).
Another structural event, which could be related to the rate-limiting step, is the passage of PQH2 from the exchange cavity to the [2Fe-2S] cluster binding niche, and as PQ in its return to the cavity after oxidation and de-protonation at the [2Fe-2S] center, through the narrow portal (Fig. 8) that connects these two compartments. Although this passage appears to be a structurally formidable task, it has not been considered in the discussions of the rate-limiting step in the entire pathway. Alternatively, it has been concluded that the rate-limiting step in the turnover of the bc1 complex is the proton coupled electron transfer from the quinol to the [2Fe-2S] cluster of the ISP (4,57).
The bn–cn branch point in cytochrome bc complexes, regulation by Fd binding
The ferredoxin/FNR-dependent cyclic electron transfer pathway as an n-side branch in the b6 f complex, as well as the presence of the unique heme cn suggested that the special function of heme cn is related to the pathway of PSI-linked cyclic electron transport, which is not present in the bc1 complex. However, biochemical (72–76), sequence (1,77) and phylogenetic (78) analysis shows that cytochrome b6 and the Rieske protein are present in the primitive nonphotosynthetic firmicutes, as is the heme cn (76). Therefore, heme cn must have another function in the “dark” metabolism of these organisms that certainly do not carry out PSI-linked cyclic electron transport. The presence of menaquinone (ol) with a potential 150 mV more negative than those of the PQ and UQ redox couples suggests the possibility of a different pathway although, as shown in the photosynthetic firmicute, Heliobacillus mobilis, redox components in the firmicute cyt bc pathway may be similarly shifted in redox potential (79). It is noted that, in spite of the presence of cyt b6 and the ISP, the polypeptide composition of the cyt bc complex in the firmicutes is very different from that of the b6 f complex: (i) the c cytochrome is distinct from cyt f and, as in Bacillus subtilis, subunit IV can be present as its N-terminal domain (75). (ii) The small hydrophobic subunits, pet G, L, M, and N are absent. Thus, the structure of the firmicute bc complex differs from that in cyanobacteria and is much simpler, although the hydrophobic core is conserved, as it is in the b6 f complex of algae and plants, and in the bc1 complex. It is suggested that the prosthetic group composition of the n-side of the cyt bc complex may dictate the details of pathways in which the complex can participate, and that the bc1, b6 f and the QCR complex of firmicutes may support three somewhat different pathways built on a similar hydrophobic core. It is also of interest to recall that the difference in redox potentials of the entire n- to p-side network in the Q cycle, extending from p-side PQ/PQH2 or semiquinone PQ couple via heme bp to heme bn–cn is small, so that an n- to p-side electron transfer pathway, reversed from the canonical Q cycle pathway, is not energetically difficult. This possibility is considered in a recent study of cyclic electron transport (80).
- Top of page
- The Problem of Proteolysis
cytochrome; DOPC/ DOPG,1,2-Dioleoyl-sn-glycero-3-phosphocholine/-phosphoglycerol
cyt b6 hemes on electrochemically positive/negative (p/n, lumen/stroma) sides of membrane
midpoint potential, pH 7
electron paramagnetic resonance
- Top of page
- The Problem of Proteolysis
Acknowledgements— We dedicate this review to our close friend and colleague, Prof. A. Hope, who contributed much to this field. This work has been supported by NIH GM-32383. We thank R. Haselkorn for suggesting the use of Anabaena sp. PCC 7120 for studies on the b6 f complex, G. Kurisu, J. Yan, and J. L. Smith for their central role in earlier stages of these studies, S. Heimann, M. Hendrich, O. Sharma, and A. Zatsman for helpful discussions, and P. Rich for a generous gift of tridecyl-stigmatellin. Much of the X-ray structure analysis was carried out at beam line SBC-19ID at the Advanced Photon Source, Argonne National Laboratory (supported by U. S. DOE W31-109-ENG-389) where S. Ginell, J. Lazarz, and F. Rotella provided important advice and technical support.
- Top of page
- The Problem of Proteolysis
- 482002) Full subunit coverage liquid chromatography electrospray-ionization mass spectrometry (LCMS+) of an oligomeric membrane protein complex: The cytochrome b6 f complex from spinach and the cyanobacterium, M. laminosus. Mol Cell Proteomics 1, 816–827., , and (
- 651992) A Q cycle type model for turnover of the bf complex under a wide range of redox conditions. In Research in Photosynthesis (Edited by N.Murata), pp. 491–494. Kluwer Scientific Publishers, Dordrecht.and (
- 812008) Structure, spectroscopy, and function of the cytochrome b6 f complex: Heme cn and n-side electron and proton transfer reactions. In Photosynthetic Protein Complexes: A Structural Approach (Edited by P.Fromme), pp. 155–179. Wiley-VCH, Weinheim., , , , and (