Transmembrane Signaling and Assembly of the Cytochrome b₆f-Lipidic Charge Transfer Complex

Abstract

Structure-function properties of the cytochrome b₆f complex are sufficiently unique compared to those of the cytochrome bc₁ complex that b₆f should not be considered a trivially modified bc₁ complex. A unique property of the dimeric b₆f complex is its involvement in trans-membrane signaling associated with the p-side oxidation of plastoquinol. Structure analysis of lipid binding sites in the cyanobacterial b₆f complex prepared by hydrophobic chromatography shows that the space occupied by the H transmembrane helix in the cytochrome b subunit of the bc₁ complex is mostly filled by a lipid in the b₆f crystal structure. It is suggested that this space can be filled by the domain of a transmembrane signaling protein.

The identification of lipid sites and likely function define the intra-membrane conserved central core of the b₆f complex, consisting of the seven trans-membrane helices of the cytochrome b and subunit IV polypeptides. The other six TM helices, contributed by cytochrome f, the iron-sulfur protein, and the four peripheral single span subunits, define a peripheral less conserved domain of the complex.

The distribution of conserved and non-conserved domains of each monomer of the complex, and the position and inferred function of a number of the lipids, suggests a model for the sequential assembly in the membrane of the eight subunits of the b₆f complex, in which the assembly is initiated by formation of the cytochrome b₆-subunit IV core sub-complex in a monomer unit. Two conformations of the unique lipidic chlorophyll a, defined in crystal structures, are described, and functions of the outlying β-carotene, a possible ’latch’ in supercomplex formation, are discussed.

Introduction

. Oxygenic photosynthesis is responsible for conversion of light energy from the sun into chemical energy stored in sugars and ATP [1]. The primary photosynthetic reactions utilizing solar energy involve the oxidation of water by the Photosystem II (PSII) protein complex to release molecular oxygen, electrons and protons (Fig. 1). On the reducing side of PSII, lipophilic plastoquinone (PQ) is reduced to plastoquinol (PQH₂), which diffuses laterally in the membrane to the cytochrome b₆f complex (cyt b₆f). The protons released by hydrolysis of H₂O contribute to the transmembrane proton electrochemical gradient, $\Delta \tilde{\mu }\mathrm{H}+$. The b₆f complex catalyzes PQH₂deprotonationand oxidation on the lumen (electrochemically positive, p) side of the thylakoid membrane, and PQ reduction and protonation on the stromal (electrochemically negative, n) side. p-side PQH₂ is oxidized by a series of sequential one electron carriers in the high potential electron transport chain consisting of: (i) the membrane bound [2Fe₂-2S] cluster in the iron-sulfur protein (ISP), usually in coordination with the p-side heme b_p of the cyt b₆ subunit of the complex, (ii) the cytochrome f heme in the p-side peripheral domain, (iii) soluble plastocyanin (PC) or cyt c₆, and (iv) the photosystem I (PSI) reaction center.

Figure 1. The electron transport chain of oxygenic photosynthesis.

Formation of the transmembrane proton electrochemical gradient coupled to the electron transport extending from H₂O oxidation to NADP⁺ reduction, in which H⁺ is translocated in the protein complexes of the PSII reaction center and cytochrome b₆f; this H⁺ gradient is utilized for ATP synthesis by the ATP synthase. PDB accession for structure data: Cyt b₆f (PDB ID 2E74), Fd (PDB ID 1EWY), ferredoxin; FNR (PDB ID 1EWY), ferredoxin- NADP⁺-reductase; PC (PDB ID 2Q5B), plastocyanin; PSII (PDB ID 3ARC) and PSI (PDB ID 1JB0), reaction center complexes.

Structure of the cyt b₆f complex

The cyt b₆f complex is a symmetric dimer [2, 3] [4] (Fig. 2A, B) that consists of eight trans-membrane polypeptide subunits. The b₆f monomer has two polytopic subunits, cytochrome b₆ (cyt b₆) and subunit IV (subIV). Cyt b₆ (~25 kDa, encoded by petB) consists of a four-transmembrane helix (TMH) bundle and a short soluble helix formed by the N-terminal 28-29 amino acids, which is associated with the n-side membrane surface. TMH helices are labeled A-D. The four TMH of cytb₆ are connected by loops that are located at the p and n-side membrane-water interfaces. SubIV (~17 kDa, encoded by petD) has three TMH (E-F) that form a p-side saddle around the four helix bundle of cyt b₆. Helix E is located in proximity to the A and B-helix of cyt b₆ while the F and G-TMHs span the four helix bundle, close to the B and C-TMH (Fig. 2C). The C-terminus of the E-TMH is separated from the N-terminus of the F-TMH by a distance of ~40 Å, which is bridged by the p-side ef-loop. This seven TMH assembly forms the polytopic core of the cyt b₆f complex. Six single TMH arising from individual subunits are arranged in the periphery around the core of the monomeric unit of the complex consisting of the cyt b₆-subIV subunits: (i) cytochrome f (cyt f, ~32 kDa, encoded by petA) and(ii) the Rieske Iron-Sulfur Protein (ISP, ~19 kDa, petC) each contain a large soluble domain on the p-side, attached to one TMH and, (iii-vi) four small (3-4 kDa) subunits, Pet G (gene petG), L (gene petL), M (gene petM) and N (gene petN), define a four-helix bundle on the periphery of each monomer of the complex, on the peripheral side opposite the inter-monomer cavity (Fig. 2D).

Figure 2. Structure of dimeric b₆f complex from M.laminosus (PDB ID 2E74).

(240,000 MW; per monomer, 8 subunits/7 prosthetic groups; 8 lipids) Subunit organization and lipid binding sites. (A) View along membrane plane showing the positions of the 8 subunits. Color code: cytochrome f (pet A), yellow; cytochrome b₆ (pet B), cyan); Rieske [2Fe-2S] protein (Pet C), orange; subunit IV/Pet D (pink), Pet G (teal), Pet L (light brown), Pet M (green) and Pet N (gray). (B) Side view of M. laminosus b₆f complex showing bound lipids, detergents and pigments. (C) Polytopic core of the b₆f complex. Cyt b₆ (4 TMH, cyan) and subunit IV (3 TMH, pink) form the core of the b₆f monomer. The cytb₆TMH form a four helix bundle. SubIV is organized around the bundle as a bipartite structure, with the E-helix separated from the F and G TMH. (D) Peripheral four helix bundle formed by the small Pet subunits, Pet G, L, M and N. In addition, including ferredoxin-NADP⁺-reductase (FNR), which may mediate electron transfer from PSI to cyt b₆f. (Fig. 1) In addition, there are four soluble subunits that associate with purified b₆f complex from plant or algal sources, but have not been seen in the crystal structures and have presumably dissociated and been lost during crystallization: the Pet P polypeptide seen in cyanobacteria, the light-harvesting LHCII chlorophyll protein kinase Stt7-STN7, the correlated phosphatase, and the PetO nuclear-encoded phosphorylatable subunit,.

Prosthetic groups

The cyt b₆f monomer has seven prosthetic groups [2, 3, 5] (Fig. 2A, B). Within the trans-membrane region, the B and D TMHs of cyt b₆ are associated with two bis-histidine ligated b-hemes (b_p and b_n, on the p and n-side). The conserved His86/His187of the B and D TMH provide axial coordination to heme b_p while His100/His202 of the B and D TMH coordinate heme b_n [2, 3, 6]. A unique covalently linked heme c_n is found to be electronically coupled to heme b_n through an axial H₂O or OH − ion, on the n-side of the complex [2, 3, 7, 8] (Figs. 2A, 3). In the absence of a second axial ligand, heme c_n is a high spin heme, with a small extinction coefficient in the visible range [9]. Cys35, a conserved residue that is found not only in the photosynthetic cyt b₆ polypeptide but also in the cyt b protein of non-photosynthetic firmicutes [10], forms a covalent thioether linkage to heme c_n (Fig. 2C). A chlorophyll-a (chl-a) and a β-carotene (β-car) molecule are also inserted between the hydrophobic TMHs (Fig. 2B). In the p-side peripheral domain, the cyt f polypeptide is covalently linked to a heme molecule (a c-type heme) while the ISP extrinsic domain binds one [2Fe-2S] cluster. The midpoint redox potentials at pH 7 of the prosthetic groups involved in electron transfer are summarized in Table 1.

Figure 3. Transmembrane heme organization within cyt b₆f.

Three hemes, b_p, b_n and c_n, are located within the b₆f hydrophobic core. Hemes b_p and b_n are axially ligated by histidines from cyt b₆. Heme c_n, a unique heme without an amino acid axial ligand, is linked to a conserved Cys35 residue of cyt b₆, and is axially ligated by a water or hydroxide molecule (red sphere).

Table 1.

Subunits of the cytb₆f complex (M. l., M.laminosus, C. r., C. reinhardtii) [39]

Subunit	MW (kDa)	MW (kDa)	pI	pI	Em7 (mV)
	M. l.	C. r.	M. l.	C. r.
Cyt f (1 heme)	32.273	31.249	6.7	8.3	+350-380
Cyt b6 (3 hemes)	24.712	24.165	9.0	8.8	(−50, bn); −50 to −150, bp; +100, Cn
ISP [2Fe–2S]	19.295	18.333	6.8	5.8	+300 - 320
SubIV	17.528	17.295	8.1	6.6	--
PetG	4.058	3.984	4.5	4.4	--
PetM	3.841	4.036	10.4	4.3	--
PetL	3.530	3.436	10.2	9.5	--
PetN	3.304	3.282	5.7	6.0	--

The inter-monomer cavity

The dimeric structure of cyt b₆f is organized around an inter-monomer cavity (30 Å high × 25 Å wide × 15 Å deep) [2, 3, 5] (Figs. 2A, B). On the n-side, the cavity is marked by the N-terminal soluble helix of cyt b₆ and on the p-side by residues from the A and D-TMH of cyt b₆. Within the lipid bilayer, the interior of the cavity is lined by amino acids from the A and D-TMH of cyt b₆, the E-helix of subIV and the ISP TMH. The cavity is considered to provide the space that sequesters the substrate PQH₂ from the membrane bilayer generated by PSII and thus to concentrate the quinol near the p-side oxidation (Q_p) site in the complex [11]. The structure around the inter-monomer cavity is stabilized through interactions between residues of cyt b₆ and subIV within the trans-membrane region. The Rieske ISP TMH is associated with the cyt f TMH of one monomer while its soluble domain crosses-over to the other monomer (Fig. 2A), where it interacts with the p-side soluble domain of cyt f and the trans-membrane regions of cyt b₆ and subIV. Further stabilization of the dimeric structure is provided by lipid molecules [12] (Fig. 2B) (and see below).

Crystal structures of the cyt b₆f complex have been obtained from the prokaryotic filamentous cyanobacteria Nostoc PCC 7120[5], M. laminosus [2] [13], and the eukaryotic green alga Chlamydomonas reinhardtii [3]. Structures of the ~250 residue cyt f peripheral domain have been obtained from C. reinhardtii [14, 15], Phormidium laminosum [16], Brassica rapa [17-19], while those of the ISP soluble domain have been solved from tryptic fragments of the ISP protein isolated from Thermosynechococcus elongates [20] and Spinacia olearcea [21]. While the trans-membrane region of the b₆f complex consists of α-helices, the N-terminal soluble domain of cyt f has an elongate β— sheet structure (Fig. 2A). The covalently linked heme f is located in the peripheral subdomain proximal to the cyt f TMH via an unusual ligation of the Tyr1 side chain to the heme Fe, along with an imidazole axial linkage from the side chain of His26 [22]. The entire cyt f soluble domain extends as an elongate 75 Å bowl shaped structure within which the ISP soluble domain is encompassed. The ISP C-terminal soluble domain is attached to the TMH through a poly-glycine hinge region [2, 3, 5] (Fig. 2A), which gives the soluble domain flexibility for motion crucial to catalysis, and consists of β–sheets separated into two smaller domains. The redox active [2Fe-2S] cluster is linked to the subdomain extrinsic to the ISP TMH. His109 and His129 side chains coordinate the outer Fe atom of the cluster while the inner Fe is linked to Cys107 and Cys127.

Sequence and structure similarity between cyt b₆f and bc₁ complexes

Cyt b₆f and bc₁ constitute the family of cytochrome bc complexes [23]. Cyt b₆f and bc₁ share several structural features. The transmembrane quinol oxidoreductase and proton pumping function of cyt b₆f is performed by the homologous cyt bc₁ complex in photosynthetic and denitrifying bacteria and in the membranes of mitochondria (Fig. 4A). Cyt b₆ has four TMH (A-D), as does the cyt b (bc₁) N-terminal domain. However, subIV has three TMH (E-G), one short of the four TMH found in the C-terminal domain of cyt b. The C-terminal 8^th helix, labeled “H” in cyt b (bc₁), is absent from subIV. Structurally, the cyt b₆ subunit is arranged as the N-terminal domain four TMH of cyt b. SubIV has a slightly altered structure (Fig. 4B) due to insertion of the chl-a molecule between the F and G TMH, and the absence of helix H from subIV.

Figure 4. Family of cyt bc complexes.

(A) Cyt b₆f and the bc₁ complex of mitochondria and anoxygenic photosynthetic bacteria constitute the cyt bc family. The transmembrane core (teal) of the bc complexes is highly conserved. The complexes catalyze quinone reduction-protonation and quinol deprotonation-oxidation, respectively, on the electrochemically negative and positive sides of the membrane to generate the proton electrochemical potential gradient. (B) Structural differences in the core: C-terminal domain of the bc₁ cyt b subunit (4 TMH, yellow) has a TMH organization that differs from subIV (3 TMH, pink) of the b₆f complex, although amino acid sequences are highly conserved.

Sequence similarity between the single helix subunits of cyt b₆f and bc₁ is limited. The ISP subunit has a significant conservation of 47-61% in the sequence only in the cluster binding sub-domain [21, 24]. Structurally, the overall fold of the ISP soluble domain and TMH are very similar between b₆f and bc₁ except in the N-terminal region that constitutes the TMH on the n-side. In b₆f, the TMH of ISP is bent while that in the bc₁ complex forms a straight α-helix [3]. There is no conservation between the sequence and structure of the b₆f cyt f and the analogous cytochrome c₁ (cyt c₁) of bc₁. Both polypeptides provide a covalently linked heme that functions as the electron acceptor for the ISP [2Fe-2S] cluster, and hence, show convergent evolution. Structurally, cyt f consists of an elongated β-sheet extrinsic domain and one TMH, while the cyt c₁ soluble domain has a globular, α-helical arrangement (reviewed recently in [25]).

Function similarity between cyt b₆f and bc₁ complexes

Cyt b₆f and bc₁ complexes act as a plastoquinol:plastocyanin and ubiquinol:cytochrome c oxidoreductase, respectively, to catalyze a “Q-cycle” [26-32]. However, the mechanism of this cycle differs in the two complexes because of the presence of the heme c_n, possibly FNR, and of the PSI cyclic pathway in the b₆f complex [10]. The enzyme FNR, found in association with the b₆f complex on the n-side [33], may donate an electron from NADPH to the Q_n-site to complete the reduction of the plastoquinone, thereby decreasing the number of p-side PQH₂- oxidation events required for the n-side reduction of PQ. The association of b₆f with FNR may be an important evolutionary adaptation with significant kinetic consequences as the p-site quinol deprotonation reaction constitutes the rate limiting step in the activity of cyt bc complexes, with an activation energy of 32 kJ/mol [34, 35]. By providing an alternate route for electron delivery to the Q_n-site, FNR may contribute to acceleration of the rate of photosynthetic electron transfer.

Differences between cyt b₆f and bc₁ complexes

(i) In addition to the linear electron transport (LET) pathway, an additional pathway, the PSI cyclic pathway operates in oxygenic photosynthesis in a feedback mechanism to transfer electrons from the reducing n-side of PSI to the quinone pool via an n-side entry to cyt b₆f and/or complex I (Fig. 1), thereby balancing redox poise and regulating the ATP level needed for carbon fixation [10]. There is no corresponding pathway in the bc₁ complex. (ii) Heme c_n, which occupies much of the n-side quinone binding niche of the bc₁ complex, is electronically coupled to heme b_n from which it is separated by 4 Å [2, 3, 5]. (iii) The small Pet subunits of cyt b₆f, PetG, L, M and N, have no homologous substitutes in the bc₁ complex [2, 3]. Instead, their position is occupied by lipids in bc₁ [36]. (iv) Cyt bc₁ also has several extrinsic polypeptides, both soluble and transmembrane, that are found only in the eukaryotic complex [37]. These subunits are absent in the prokaryotic bc₁ [38] and the photosynthetic b₆f complexes [2, 3]. (v) The b₆f complex contains three more prosthetic groups than bc₁. While both complexes have hemes b_p and b_n in the transmembrane region and a [2Fe-2S] cluster and a covalently linked heme in the soluble domains, the unique heme c_n, chl-a and β-car are not found in the bc₁ complex [39]. (vi) The architecture of the Q_p and Q_n sites also differs between the complexes due to the presence of a chl-a and the unique heme c_n in the b₆f complex. On the p-side, the heterocyclic chlorin ring of the chl-a molecule is located between the F and G TMH of subIV (Fig. 2B) while its long phytyl-tail passes into the portal of quinone entry that leads to the Q_p-site in b₆f [5, 11]. This portal is formed by residues from the C-helix of cyt b₆ and the F-helix of subIV and measures 12-15 Å in cross-section. The absence of chl-a from the bc₁ complex enlarges the Q_p-portal, which may have effects on the residence time of the substrate quinol within the Q_p-site, thereby providing a major difference between structure and kinetics on the p-side of cyt b₆f and bc₁. On the n-side, heme c_n (Fig. 2C) is located at the Q_n-site, with its open axial position facing the inter-monomer cavity. Crystallographic studies have implicated this open axial position as the quinone binding site in b₆f [13]. Compared to the Q_n-site of bc₁ that consists of an amino acid environment leading to heme b_n, the Q_n-site of b₆f provides greater access to the substrate quinone from the inter-monomer cavity. This structural difference has an important implication in reducing the selectivity and efficiency of inhibitor binding to the Q_n-site (discussed below), again providing for major structural, kinetic and equilibrium related differences between b₆f and bc₁. In summarizing the differences between the two sets of complexes, it can be said that the b₆f complex is not a trivially modified bc₁ complex.

Lipids associated with the b₆f complex

Membrane proteins depend on internal integral lipids for structural stability and biological activity [40-42]. The role of lipids in stabilizing the structure of membrane proteins, or even inducing structural changes within proteins has been investigated by methods that are sensitive to changes in secondary and tertiary structure, thermal stability assays using, for example, enzymatic activity, that provide information about the functional dependence of the protein on lipid content [40, 43].

It is now well documented that successful crystallization of a large number of membrane proteins depends on the presence of stabilizing lipids, as in the case of the photosynthetic light harvesting complex [44], bacteriorhodopsin [45], GPCRs [46] and aquaporin [47]. High resolution crystal structures of membrane proteins provided evidence for specific lipid-protein interactions, as deduced from the presence of well-defined lipid binding sites occupied by ordered lipids, both on the surface and within folded polypeptides. The class of membrane proteins that represents the largest number of well-defined lipid binding sites observed structurally is that of the hetero-oligomeric complexes that consist of distinct polypeptide subunits [48, 49]. Membrane protein complexes are assembled from protein components of different sizes and amino acid composition. The problem of packing the components together to form a stable membrane protein complex is solved by the incorporation of lipid molecules, which interact with the surrounding protein environment, mainly through Van der Waals interactions [50, 51]. Due to the non-specific nature of these interactions, it may be possible to substitute a natural lipid with a synthetic lipid without affecting biological function [52-54]. During the process of purification, natural lipids may also be replaced by detergent molecules that serve as synthetic structural and functional analogs of lipids [48]. Due to the smaller volume occupied by a single tail, detergents tend to form micelles, and not bilayers like lipids [40]. The physico-chemical similarity between the properties of lipids and detergents has implications for the identification of lipid binding sites [48, 49]. The presence of a crystallographically well defined site with a bound detergent molecule is considered to mark the position of a natural lipid. Weakly bound lipids may be replaced by detergent molecules, especially at the surface of membrane proteins. Therefore, the analysis of lipid binding sites in membrane proteins is inclusive of sites occupied by natural lipids, synthetic lipids as well as ordered detergent molecules.

Prior to elucidation of the cyt b₆f crystal structure, a role of lipids in the structure had been inferred through biochemical studies [55, 56]. Successful crystallization of the cyanobacterial cyt b₆f complex to obtain highly diffracting crystals required the addition of a synthetic lipid to the purified and delipidatedb₆f preparation. Addition of the neutral lipid dioleoylphosphatidylcholine (DOPC) improved diffraction quality of b₆f crystals from 11 Å to 3 Å [2, 50]. Purification of the b₆f complex from the eukaryotic alga C. reinhardtii for successful crystallization was carried out under mild conditions of metal-affinity chromatography that did not remove lipids from the complex [3]. Elucidation of the structure of cyt b₆f from C. reinhardtii showed the presence of a total of 4 lipids and 4 detergent bound sites within the transmembrane domain, in addition to the hydrophobic photosynthetic pigments chl-a and β–carotene (Fig. 2A, B).

The use of hydrophobic chromatography allowed manipulation of lipid content in the eight lipid binding sites of cyt b₆f (Fig. 2B). The finding that the hydrophobic space occupied by the eighth transmembrane helix of the respiratory and anoxic photosynthetic cytochrome bc₁ complex is occupied by a lipid and a chlorophyll in the b₆f complex raises the question of the pressure in evolution that led to this change [12, 57]. The question is posed of the function of the lipid substitution in relation to the evolutionary change between the eight and seven helix structures of the cyt b polypeptide [57]. Based on the known n-side activation of the light harvesting chlorophyll protein kinase by p-side plastoquinol [58], one possibility is that the change was directed by the selective advantage of p- to n-side trans-membrane signaling functions in b₆f, with the lipid either mediating this function, or substituting for the TMH of a signaling protein lost in crystallization.

Assembly of the hetero-oligomeric cyt b₆f complex

As discussed above, the active dimeric cyt b₆f complex consists of at least 8 distinct transmembrane gene products (cyt f, cyt b₆, Rieske ISP, subIV, PetG, L, M and N), along with 7 tightly bound prosthetic groups (hemes f, b_p, b_n and c_n, [2Fe-2S] cluster, chl-a and β-car) (Figs. 2A, B). The genes for the subunits do not constitute a single operon. The cyt b₆ gene (petB) and the subIV gene (petD) are under the genetic control of a single promoter, as the petBD operon [59, 60]. In prokaryotes, the ISP gene (petC) and the cyt f gene (petA) are also organized into an operon, the petCA operon. Hence, the transcription of the four major subunits is expected to be genetically co-ordinated. However, the four peripheral subunits, Pet G, L, M and N, are not encoded by genes that are within an operon. Genetic control and coordination of the Pet G, L, M and N synthesis is not well understood.

Mutagenesis studies have provided information about the sequence of events that lead to the assembly of the dimeric cyt b₆f complex [61-69]. Heme binding has been suggested to be an important event in stabilization of the cyt b₆f component polypeptides. In the absence of bound heme moieties, the cyt b₆ and cyt f polypeptides are highly sensitive to proteases [63, 64]. An analysis of the cyt b₆f crystal structures shows that the cyt b₆ hemes b_p and b_n are involved in structural stabilization to the cyt b₆ polypeptide. Axial coordination to the hemes b_p and b_n is provided by His86/His187 and His100/His202 respectively of the cyt b₆ polypeptide, which leads to its folding into a compact four helix bundle [2, 3]. Heme-mediated stabilization of the cyt f subunit occurs via an unusual ligation of the heme via the Tyr1 residue side-chain of the mature cyt f polypeptide [18]. Crystal structures of the isolated cyt f extrinsic domain and of the intact dimeric cyt b₆f complex show virtually no structural differences between the cyt f extrinsic domain. Hence, heme binding is an important event in the folding and stabilization of the subunits of the cyt b₆f complex.

The hetero-oligomeric nature of the b₆f complex requires the presence of lipids to enhance and strengthen protein-protein interactions between non-identical subunits [52]. Lipids present on the surface and between polypeptide subunits provide cross-linking interactions that contribute to stability of the complex. As discussed elsewhere [12], crystal structures of b₆f show well-defined lipid, detergent and lipidic pigment binding sites. The organization of lipidic molecules within the b₆f complex provides clues to the process of assembly. Given the close interactions between the distinct polypeptides and the integral association of prosthetic groups and lipids with the folded polypeptides, it is important to determine the mode of cyt b₆f assembly in the thylakoid membrane, both in terms of the order in which various subunits are assembled and the stage(s) of insertion of prosthetic groups.

Assembly of a functional cyt b₆f complex dimer would require coordination in time of synthesis of the various subunits in proximity of the synthesized products to ensure that all necessary polypeptides are available to assemble into the functional complex. On the basis of the genetic organization of cyt b₆f and crystal structures of the purified dimeric complex from the cyanobacteria M. laminosus (PDB IDs 1VF5, 2E74) [2, 13] and Nostoc PCC 7120 (PDB 2ZT9) [5], and C. reinhardtii (PDB ID 1Q90) [3], a model is proposed to explain the sequence of events that lead to the formation of the dimeric cyt b₆f complex in thylakoid membranes. The first step is considered to involve the transcriptional activation of the petBD operon that encodes the polytopic cyt b₆ and subIV subunits of b₆f that form the core. Following transcription, the mRNA is translated into the cyt b₆ and subIV polypeptides that undergo co-insertion into the membrane to form the polytopic monomeric core of the cyt b₆f complex (Fig. 5, step 1). Dimerization of the polytopic monomer is expected to follow its assembly. As seen in the crystal structures of the b₆f complex, lipids, and detergents that mark physiological lipid binding sites act as cross-linkers to stabilize monomer-monomer interactions at the interface of the dimer. Three lipid sites mediate inter-monomer interactions, viz., (i) p-side lipid, mimicked by the synthetic lipid DOPC (Fig. 5, step 2, DOPC lipid shown as red sticks) in the M. laminosus cyt b₆f structure (PDB ID 1VF5) located at the opening of the inter-monomer cavity, and (ii, iii) two n-side UDM detergent occupied sites (Fig. 5, step 1, UDM detergent shown as red/white sticks) that interact with the cyt b₆ N-terminus of one monomer and the cyt b₆ D TMH of the other monomer in M. laminosus cyt b₆f (PDB ID 2E74) (Fig. 5, step 3). The ISP and cyt f polypeptides are expected to be co-inserted into the thylakoid membrane to form an ISP-cyt f sub-complex (Fig. 5, step 4). The interaction between the two polypeptides is strengthened on the n-side by the native acidic sulfolipid that interacts with basic residues from the trans-membrane helices of ISP and cyt f. On the p-side, the ISP-cyt f interaction is mediated by another lipid binding site. A detergent and a photosynthetic pigment eicosane were modeled into this site in crystal structures of the b₆f complex from cyanobacteria (PDB IDs 2E74, 2ZT9) and C. reinhardtii (PDB ID 1Q90). Van der Waals interactions stabilize the interaction between lipid in the p-side site and the residues of the ISP and cyt f TMH. This ISP-cyt f sub-complex then interacts with the cyt b₆-subIV polytopic core nucleus to form a cyt b_6-subIV-ISP-cyt f sub-complex. As discussed elsewhere, the acidic sulfolipid and the ligand in the p-side lipid site contribute to the association through Van der Waals interactions with residues from the polytopic core. Biochemical evidence supports the existence of an ISP-cyt f sub-complex in spinach [70]. The proposed model explains the existence of these smaller complexes as a consequence of proximity between the genes that encode these subunits, which allows for coordination in transcription, translation, membrane insertion and assembly.

Figure 5. Proposed model for assembly of the cyt b₆f complex.

(Step 1) The first step involves the assembly of the monomeric polytopic core of cyt b₆ and subIV. (Step 2) The monomeric polytopic core then undergoes dimerization, mediated by the cross-linking interactions of lipids (DOPC, shown as red sticks; UDM shown as red/white sticks), which leads to the formation of the dimeric cyt b₆-subIV polytopic core (Step 3). Formation of the ISP-cyt f sub-complex takes place via the stabilizing interactions of lipids (Step 4). Assembly of the PetGLMN sub-complex takes place around the β-carotene molecule (Step 5). Interaction of the PetGLMN sub-complex with the core polytopic core takes place via the β-carotene and lipids (sulfolipid, DOPC, MGDG). Alternatively, formation of an ISP-cyt f-PetGLMN sub-complex may take place prior to interaction with the dimeric core (Step 6). (Step 7) Interaction of the peripheral sub-complex with the dimeric core would lead to the formation of the fully assembled, functional dimeric cyt b₆f complex.

The four small subunits of the cyt b₆f complex, Pet G, L, M and N, are organized into a four helix bundle in the structure of the b₆f complex [2, 3, 5, 13]. Pet L, M and N form a β-car binding site. The sequence conservation of these subunits is low. Pet G, the subunit with the highest degree of sequence conservation does not interact with the β-car [12, 71]. Since these subunits are encoded by genes that are separated by large distances, the assembly process would require a long-range communication for coordination between genetic transcription, translation, membrane insertion and assembly. The structure of the b₆f complex shows that on the n-side, the extrinsic portion of these subunits consists of basic amino acid residues [11, 72]. Based on the cis-positive rule [73] and the short length of the Pet subunits helices that span the hydrophobic membrane only once, it has been suggested that they are inserted spontaneously into the membrane [72]. It is expected that organization of the Pet subunits around the β-car molecule leads to the formation of the four helix bundle (Fig. 5, step 5). Assembly of the PetGLMN sub-complex may take place independently on the periphery of the cyt b₆f polytopic core. Mutagenesis studies have shown that Pet G and Pet N, which show relatively high sequence conservation, are essential to the stability of the dimeric b₆f complex [71, 74, 75].

The crystal structure of the cyt b₆f complex shows a clear separation between the polytopic core domain and the peripheral domain that consists of single helix subunits [12]. It is noted that the PetGLMN sub-complex may interact with the ISP-cyt f sub-complex (Fig. 5, step 6), leading to the formation of a peripheral sub-complex that interacts with the polytoptic core, which eventually forms the hetero-oligomeric, fully functional cyt b₆f complex (Fig. 5, step 7).

Cyt b₆f and super-complex formation

Elucidation of the structural organization of cyt bc complexes provided an understanding of the function performed by these quinone oxidoreductases in energy transduction. Sequence conservation of the transmembrane core subunits [6] was found to extend to the structure cyt bc complexes, between the cyt b polypeptide of the bc₁ complex and the cyt b₆/subIV polypeptides of the cyt b₆f complex (discussed above). A significant and conserved structural feature relates to the organization of the p-side quinol oxidation (Q_p) site. It has been demonstrated that the deprotonation of the semiquinone intermediate within the Q_p-site is mediated by the Glu residue of the conserved PEWY sequence located on the ef-loop [76-78]. In the cyt bc₁ complex (PDB ID 3CX5) [79], the ef-loop is inserted between the F and G TMH of the cyt b subunit (Fig. 6, left panel). Stabilization of the Q_p-site architecture depends on the 8^th C-terminal “H” TMH of the cyt b subunit in the bc₁ complex. The helix is inserted in the space between the F and G helices and hence, stabilizes the inter-helix space that is required for the insertion of the ef-loop. A very similar structural arrangement of the Q_p- site is observed in the b₆f complex (Fig. 6, right panel). However, it has been noted previously that an unusual evolutionary structural difference between the cyt b₆f and bc₁ complexes is seen in the organization of the cyt b polypeptide [12, 57]. The ‘H’ TMH, which performs an important role of stabilizing the niche for the catalytically important ef-loop, is absent from the photosynthetic cyt b₆f complex. As demonstrated, the niche of the missing H TMH is occupied by a lipid and a chl-a molecule in the cyt b₆f complex.

Figure 6. Organization of the Q_p-site in cyochrome bc complexes.

The ef-loop bears the conserved PEWY sequence, whose Glu residue is involved in the second deprotonation reaction (highlighted in reaction sequence) of the substrate within the Q_p-site. Left panel: In the cyt bc₁ complex (PDB ID 3CX5), the ef-loop is inserted between the F and G transmembrane helices of the 8 helix cyt b polypeptide (shown in block diagram at the bottom). The space between the F and G TMH is stabilized by the H TMH of cyt b. Right panel: In the cyt b₆f complex (PDB ID 2E74), the ‘H’ TMH is absent from the subIV polypeptide (shown as block diagram). This niche is occupied by a lipid and a chlorophyll-a molecule.

Protein-to-lipid substitution: Structure-driven evolutionary modification

Why should a transmembrane helix be replaced by a lipid and a lipid-like chl-a molecule? A possible explanation may lie in the pathway of state transitions, which is unique to photosynthesis (Fig. 7) [80]. Photosynthetic state transitions balance the redox poise of the plastoquinone (PQ)/plastoquinol (PQH₂) pool within the thylakoid membrane. The Photosystem-II (PSII) reaction center utilizes solar energy to catalyze the reduction of PQ to PQH₂ (Fig. 7A). The reduced PQH2 moiety then diffuses to the cyt b₆f complex, where it undergoes oxidation within the Q_p-site. Based on studies performed on the cyt bc₁ complex, quinol oxidation within the Q_p-site of the cyt b₆f complex is expected to constitute the rate limiting step in the photosynthetic electron transport chain. As a result, PQH₂/PQ redox poise shifts to a predominantly reduced state (Fig. 7B). To restore balance, an enzyme identified as the LHCII kinase Stt7 in C. reinhardtii, is activated [81, 82]. The LHCII kinase phosphorylates the accessory antenna LHC molecules associated with the PSII reaction center, causing their dissociation and migration to photosystem I (PSI). In the absence of LHC molecules, the efficiency of light harvesting and catalysis of PQ reduction by PSII is significantly diminished (Fig. 7C). PQH₂ consumption by the cyt b₆f complex is not affected, and hence, redox balance is restored in the PQ/PQH₂ pool. Mutagenesis experiments have shown that the cyt b₆f complex acts as the sensor of the quinone pool redox state [58]. Binding of PQH₂ within the Q_p-site generates a signal that activates the LHCII kinase. Biochemical studies of the Stt7 kinase in C. reinhardtii show that the kinase has a membrane spanning domain along with an extrinsic domain on n- side of the thylakoid membrane [81, 82]. Catalytic activity of the LHCII kinase resides in the extrinsic domain, and is activated by PQH₂ binding within cyt b₆f on the p-side of the membrane. The kinase activating signal has to be transduced across the hydrophobic bilayer. The exact mechanism by which PQH₂ binding generates the signal and the identity of the signal are presently unknown.

Figure 7. Photosynthetic state transitions.

(A) The photosystem II (PSII) reaction center complex utilizes light energy harvested by accessory light harvesting complex II (LHC) molecules to reduce plastoquinone (PQ) to plastoquinol (PQH₂). PQH₂ undergoes diffusion to the Q_p-site (blue circle) of the cyt b₆f complex (brown cartoon), where it undergoes oxidation. (B) Slow PQH₂ oxidation by cyt b₆f causes PQH₂ accumulation in the thylakoid membrane, which activates the LHCII kinase (shown in red). (C) In the phosphorylated state, LHC molecules migrate to photosystem I (PSI), thereby reducing the efficiency of PQ reduction by PSII. As PQH₂ consumption by cyt b₆f is not affected, a redox balance is restored in the quinone pool, which inactivates the LHCII kinase (shown in gray).

LHCII kinase activation: Putative mechanism

A comparative analysis of the cyt b₆f crystal structures from cyanobacteria (PDB IDs 2E74, 2ZT9) and C. reinhardtii (PDB ID1Q90) shows an interesting structural feature that may be relevant to transmembrane signal transduction to activate state transitions. The chl-a chlorin-ring is inserted between the F and G TMH of subIV (Fig. 8). The phytyl-tail of chl-a passes through the portal that leads into the Q_p-site for substrate binding. The phytyl-tail of the chl-a has different conformations in the crystal structures of cyt b₆f obtained from cyanobacteria (PDB IDs 2E74, 2ZT9) and C. reinhardtii (PDB ID 1Q90). In the cyanobacterial cyt b₆f complex, the chl-a phytyl tail is wrapped around the F TMH while in the cyt b₆f structure from C. reinhardtii, the phytyl-tail is located proximal to the C TMH (Fig. 8). The alternate positions indicate potential flexibility in the chl-a phytyl-tail. It is proposed that upon binding of the natural substrate plastoquinol-9 within the Q_p-portal of cyt b₆f, the 45 carbon long isoprenoid tail of plastoquinol-9 molecule displaces the chl-a phytyl-tail, thereby modifying the interaction between chl-a and the ‘F’ TMH (Fig. 9). It is proposed that changes in the chl-a/F TMH interaction generate a signal that is transduced to the n-side of the cyt b₆f complex, either directly through the ‘F’ TMH or by a combinatorial mechanism that involves the ‘F’ and lipid substituting for the ‘H’ TMH. The ‘F’ TMH is the prime candidate for signal transduction as it provides a direct physical connection between the substrate binding Q_p-site on the p-side and the n-side surface of the cyt b₆f complex.

Figure 8. Alternate conformations of the chl-a phytyl-tail in cyt b₆f.

The chl-a chlorin-ring is inserted between the F and G TMH of subIV (light pink) while the tail is inserted between the F TMH and the C TMH of cyt b₆ (pale cyan). In the crystal structure of cyt b₆f (PDB ID 2E74) from the cyanobacterium M. laminosus, the chl-a phytyl-tail (green) is wrapped around the F TMH while in the structure obtained from C. reinhardtii, the chl-a phytyl-tail (red) is located proximal to the C TMH.

Figure 9. Proposed mechanism of chl-a mediated signal transduction in the cyt b₆f complex (PDB ID 2E74) for activation of photosynthetic state transitions.

(Step 1) Binding of the natural substrate plastoquinol (black/red sticks) within the Q_p-site displaces the chl-a phytyl-tail (displacement shown as broken line with arrows) due to interaction with the long isoprenoid tail of the palstoquinol molecule. The displacement event generates a signal that is transduced to the n-side of the complex either (Step 2) via the F TMH or (Step 3) through a combination of the F TMH and the n-side lipid (yellow) present between the F and G TMH. The LHCII kinase is shown (gray) in cartoon format.

It is significant to note that no structure is available for the isolated, purified LHCII kinase. Biochemical evidence is not available to determine the binding site for the kinase on the cyt b₆f complex. It is possible that the lipid bound between the ‘F’ and ‘G’ TMH of cyt b₆f actually represents the LHCII kinase binding site. Crystallographic evidence can be obtained from cyanobacterial crystal structures of the cyt b₆f complex [5, 13] in which DOPC, a non-physiological lipid, is able to bind in the niche. Hence, the lipid binding site between the ‘F’ and ‘G’ TMH is not expected to have a very high affinity for lipids. This inference is further supported by the crystal structure of the cyt b₆f complex obtained from C. reinhardtii [3]. A long, tubular electron density is observed in the binding niche between the F and G helices, which may belong to lipid present at a low occupancy (Fig. 10). Based on these observations of an exchangeable lipid, it is proposed that the niche formed between the F and G TMH may represent the binding site for the LHCII kinase (Fig. 11). In this model, the presence of the lipid may represent a state in which the cyt b₆f- LHCII kinase association has not been established. Formation of the cyt b₆f-LHCII kinase super-complex would be accompanied by the replacement of the n-side lipid by the transmembrane domain of the LHCII kinase. The LHCII kinase would interact with the Q_p-site and the F TMH of cyt b₆f. According to the proposed model, replacement of the H TMH of the cyt b polypeptide is explained by the need to generate the binding site for the LHCII kinase.

Figure 10. Potential lipid binding site in cyt b₆f complex of C. reinhardtii.

The cyt b₆f structure (PDB ID 1Q90) was obtained in the presence of native lipids. Residual electron density is shown as green mesh (highlighted by broken black line) between the Fo-Fc map (3.0 σ). The electron density corresponds to the n-side lipid in the cyanbacterial cyt b₆f complexes from M. laminosus (PDB ID 2E74) and Nostoc PCC 7120 (PDB ID 2ZT9). The electron density map was obtained from the Electron Density Server [96].

Figure 11. The lipidic mechanism of cyt b₆f-LHCII kinase super-complex formation.

It is proposed that the lipid DOPC (yellow/red/blue sticks) is bound weakly in the niche formed between the F and G TMH, which represents the LHCII kinase (shown as cartoon. gray) binding site. Recruitment of the LHCII kinase to the cyt b₆f complex and binding results in the replacement of the lipid from the inter-helix niche.

Structure-linked role of β-carotene in cyt b₆f complex

The model of LHCII kinase binding and activation assigns a structural role to the photosynthetic pigment chl-a, instead of a photochemical function. This is a unique role that is not commonly observed. In the cyt b₆f complex, in addition to the chl-a molecule, one β–car is inserted into the periphery of the complex, in between the peripheral Pet subunits, as described elsewhere [2, 3] and discussed above. Biological systems that harvest light, such as photosystems and light harvesting complex proteins, utilize chlorophyll molecules to capture light and initiate photochemistry. However, in the excited triplet state, the chlorophyll molecule has the potential to transfer energy to molecular oxygen, leading to the formation of harmful reactive oxygen species [83]. β–Car molecules are utilized as quenchers in such biological systems to dissipate energy from the potentially harmful excited triplet state of chlorophyll. However, the separation between the chl-a and the β– car molecule within the cyt b₆f monomer is 14 Å, which is significantly larger than a 4 Å distance that would allow efficient wave-function overlap [84]. It has been suggested that the energy transfer may take place between the chl-a and β–car of cyt b₆f via an oxygen-channel [85, 86]. However, experimental demonstration of such a channel remains elusive. As an alternative function, the β–car molecule is thought to perform two structural roles [12, 57]. The β–car molecule may mediate assembly of the heterooligomeric cyt b₆f complex by providing a scaffold for the organization of the PetGLMN sub-complex. Additionally, it has been suggested that the β–car molecule may function as a molecular antenna that mediates the search for other photosynthetic complexes, such as PSI, and then serves as a latch to mediate the formation of a cyt b₆f-PSI super-complex. The isolation of an enzymatically functional cyt b₆f-PSI super-complex has recently been reported from C. reinhardtii thylakoid membranes [87]. While the models proposed in the present discussion suggest structural functions for the photosynthetic pigments chl-a and β–car, the problem of the large 14 Å separation between the pigment molecules in the cyt b₆f complex still remains. In the absence of a β–car mediated pathway to channel energy from the excited chl-a molecule, the cyt b₆f complex is prone to generate reactive harmful species that may cause oxidation of biological molecules. It is expected that a mechanism for energy dissipation from the chl-a moiety of cyt b₆f must be operational under in-vivo conditions of photosynthetic electron transfer.

Structural studies of the cyt b₆f complex have so far been performed in isolation with crystallization grade purity [50]. While crystallographic studies of the isolated cyt b₆f complex have yielded valuable information, it has not been possible to study the interactions of the cyt b₆f complex with other membrane proteins, such as the photosystems. Within the thylakoid membrane, photosystems, cyt b₆f and LHC complexes do not operate in isolation. Interaction between the large hetero-oligomeric membrane bound complexes mediates physiologically relevant photochemistry and electrochemistry. On these lines, it is possible that the chl-a molecule of the cyt b₆f complex is quenched by a β-car from another membrane bound photosynthetic protein, such as a photosystem or an LHC molecule. Such an interaction would require close interaction between cyt b₆f and other membrane bound proteins. With the elucidation of the existence of a cyt b₆f-PSI super-complex [87], evidence has been presented for physical association between the hetero-oligomeric photosynthetic complexes. An analysis of the crystal structures of Photosystems I [88] and II [89, 90] and of the light harvesting complex II [91] shows that peripheral carotenoids are ubiquitous in these complexes (Fig. 12). Hence, it is possible that the formation of a super-complex of the cyt b₆f complex not only mediates metabolic processes, but also provides the necessary pathway for quenching of the chl-a excited state within the b₆f complex.

Figure 12. Arrangement of β–carotene in the major photosynthetic integral membrane protein complexes (A) photosystem II (PDB ID 3ARC), (B) photosystem I (1JB0) and light harvesting complex II (PDB ID 2BHW).

The β–car molecules are shown as yellow sticks while the polypeptides are represented as thin green ribbons. The β-car molecules are located on the periphery of the complexes, providing potential interaction sites for the chl-a of cyt b₆f. All three complexes are viewed along the normal to the membrane plane. For simplicity, other prosthetic groups have been omitted.

Selectivity (and the lack thereof) in the Q_n-site of cyt b₆f complex

A useful tool in studies of electron transfer through the hemes and [2Fe-2S] cluster prosthetic groups of cyt bc complexes has been the use of specific quinone analog inhibitors [92-94]. Inhibitors approximately simulate the binding properties of the natural substrate to occupy the Q_p and Q_n sites of quinol oxidation and quinone reduction on the p and n-sides of cyt bc complexes. In the presence of inhibitors, electron transfer is inhibited through the cyt bc complexes. Specific inhibitors have been designed to block defined pathways of electron transfer. As a result, characterization of inhibitors has been performed according to their specific role in blocking the process of electron transfer. With regard to the cyt bc₁ complex [94], class I inhibitors include stigmatellin and 5-nundecyl-6-hydroxy-4,7-dioxobenzothioale (UHDBT) that bind within the Q_p-site and interact with the ISP subunit. The binding of class I inhibitors has been characterized to occur in a ‘pocket’ removed from heme b_p and within H-bond distance of an ISP histidine residue. On the other hand, class II inhibitors include myxothiazole and β-methoxyacrylate stilbene (MOA-stilbene) and famoxadone that bind close to heme b_p and do not interact with the ISP protein significantly, with the exception of famoxadone. Together, the class I and II inhibitors comprise the p-side quinone analog inhibitors that prevent quinol oxidation within the Q_p-portal of the cyt bc₁ complex. On the n-side of the cyt bc₁ complex, one of the most well-characterized and potent inhibitors is antimycin-A. The inhibitor antimycin-A prevents the n-side reaction, i.e., reduction of quinone to quinol bound at the Q_n-site of the cyt bc₁ complex. Distinction of quinone analog inhibitors as p and n-side specific agents is possible from crystallographic studies of cyt bc₁ complex [38, 94, 95]. Crystallographic studies of the purified cyt b₆f complex from the cyanobacterium M. laminosus present a different situation [13]. Tridecyl-stigmatellin (TDS), an analog of the cyt bc₁ p-side inhibitor stigmatellin was used for cocrystallization studies. As expected from previous studies, TDS was found to bind at the Q_p-site within the cyt b₆f complex. Surprisingly, electron density for the chromone ring of TDS was also observed on the n-side, bound axially to the unique heme c_n. Binding of TDS at the Q_n-site of cyt b₆f indicates that selectivity of inhibitor binding at the Q_n-site may be less well defined in cyt b₆f compared to the cyt bc₁ complex.

An analysis of the polypeptide environment around the Q_n-site of the cyt b₆f complex provides an explanation for the limited selectivity of inhibitor binding (Fig. 13). The Q_n-site in the cyt b₆f complex consists of a heme b_n/c_n couple [13]. The presence of heme c_n extends the binding site of the substrate quinone farther away from the protein environment, into the inter-monomer cavity of the cyt b₆f dimer. Using the TDS molecule bound axially to heme c_n as a marker for the natural quinone, it is found that only six amino acids lie within an interaction distance (i.e., within 4.0 Å) of the TDS chromone ring (Fig. 13A). Val26 of cyt b₆ and Ala31, Asp35, Leu36 and Phe40 interact with the TDS chromone ring through Van der Waals interaction. Arg207 of cyt b₆ forms the only selective hydrogen bond interaction with the TDS chromone ring. In contrast, in the cyt bc₁ complex (PDB ID 1KB9) [36], the ring of the ubiquinone-6 molecule bound at the Q_n-site interacts with seven residues from the cyt b polypeptide within a distance of 4.0 Å. Tyr16, Gln22, Trp30, Ser206 and Asp229 form polar interactions with the quinone ring while Leu201 and Met221 interact via Van der Waals interactions (Fig. 13B). Hence, selectivity is larger in the cyt bc₁ complex due to the larger fraction of selective polar interactions. This structural feature may also explain the observation that no quinone analog inhibitor for the cyt b₆f Q_n-site has been found to be comparable in efficiency to antimycin bound at the Q_n-site of cyt bc₁ [92]. It is inferred that binding of ligands and quinone analog inhibitors at the Q_n-site is less selective in the b₆f complex than in the bc₁ complex, mainly due to fewer interactions.

Figure 13. Binding of ligands to the Q_n-site of cyt bc complexes and the role of the protein environment in selectivity.

(A) In the cyt b₆f structure (PDB ID 2E76) obtained from the cyanobacterium M. laminosus, the majority of the interactions to the quinone analog inhibitor TDS (green/red ball and stick model) are contributed through non-specific Van der Waals interaction. (B) On the other hand, binding of the natural ligand ubiquinone-6 (green/red ball and stick model) to the cyt bc₁ complex (PDB ID 1KB9) Q_n-site is mediated by polar interactions. For simplicity, the hydrophobic tails of TDS and ubiquinone-6 have been omitted.

Highlights.

• We compare cytochrome b₆f and bc₁ crystal structures to identify novel lipid functions.

• The role of lipids in mediating assembly of the b₆f complex is discussed.

• A lipid mediated transmembrane signaling mechanism is suggested for photosynthetic state transitions.

Abbreviations

β-Car

β-carotene

Chl-a⁻

chlorophyll-a

Cyt

cytochrome

Em7

midpoint oxidation-reduction potential at pH 7

$\Delta \tilde{\mu }\mathrm{H}+$

trans-membrane proton electrochemical potential gradient

n, p-sides

electrochemically negative and positive side of the membrane

PDB

protein data bank

PC

plastocyanin

PQ

plastoquinone

PS

photosystem

subIV

subunit IV of the b₆f complex

TMH

transmembrane helix