Present†Current address: OX11 0DE, UK, Diamond Building, Harwell Science and Innovation Park, Dietcote, Oxfordshire, UK, Diamond Light Source Co., Ltd., Electronic Biological Imaging Center.
The reaction center light-harvesting complex 1 (RC-LH1) is the core photosynthetic component of purple phototrophic bacteria. We introduced two cryo-electron microscopy structures of the RC-LH1 complex from Rhodopseudomonas palustris. The 2.65-Å resolution structure of the RC-LH114-W complex consists of 14 subunit LH1 loops surrounding RC, which is interrupted by protein W, while the complex without protein-W is completely RC composition surrounded by RC. Closed 16 subunit LH1 loop. The comparison of these structures provides insights into the dynamics of quinone in the RC-LH1 complex, including previously undetermined conformational changes when binding quinone at the RC QB site, as well as the location of auxiliary quinone binding sites, which help Pass them to RC. The unique structure of the W protein prevents the closure of the LH1 loop, thereby creating a channel for accelerating the quinone/quinolone exchange.
The energy provided by photosynthesis can sustain almost all life on earth, and it has great potential for solar biotechnology. While promoting global photosynthesis, purple phototrophic bacteria also exhibit various energy modes and metabolic capabilities. They can avoid photosynthesis and grow as heterotrophic bacteria in the dark, can fix nitrogen and carbon dioxide, produce hydrogen, and degrade aromatic compounds (1-3). In order to provide energy for these processes, light must be quickly and efficiently converted into chemical energy. This process starts when the light-trapping antenna complex absorbs light and transfers the trapped energy to the reaction center (RC), thereby starting charge separation (4 – 7). The basic unit of photosynthesis in purple phototrophic bacteria is composed of type 2 RC, surrounded by light-harvesting complex 1 (LH1), forming the RC-LH1 core complex. LH1 is formed by an array of curved αβ heterodimers, each of which binds two bacterial chlorophyll (BChl) a molecules and one or two carotenoids (8-12). The simplest LH1 antenna consists of 16 or 17 αβ heterodimers encircling RC (9-13) in a closed loop, but in other core complexes, transmembrane peptides interrupt the continuity of the surrounding LH1 , Thereby promoting the quinol/quinone diffusion between RC and cytochrome bc1 complex (11, 13-15). The purple phototrophic plant Rhodopseudomonas (Rps.) is a model organism that can understand the energy and electron transfer that supports photosynthesis. The first crystal structure of Rps. The model of the palustris RC-LH1 complex is RC, surrounded by 15 heterodimeric LH1 loops, which are interrupted by an unknown protein called “Protein W” (14). Protein-W was subsequently identified as RPA4402, which is an uncharacterized 10.5kDa protein with three predicted transmembrane helices (TMH) (16). We propose to rename the rpa4402 gene encoding protein W to pufW to be consistent with the nomenclature used for genes encoding RC-L, M (pufL, pufM) and LH1α, β (pufA, pufB) subunits. Interestingly, protein-W is only present in about 10% of RC-LH1, revealing Rps. palustris produces two different RC-LH1 complexes. Here, we report the high-resolution cryo-EM (cryo-EM) structures of two core complexes, one with protein W and 14 αβ heterodimers, the other without protein W and a closed 16 Heterodimer LH1 loop. Our structure represents a step change in the understanding of the RC-LH1 complex of Rps. palustris, because we have analyzed the homogeneous population of each variant and have sufficient resolution to clearly assign each peptide and bound pigments and related lipids and quinones. The comparison of these structures shows that the three TMH proteins-W that have not been found in any other RC-LH1 complex so far generate a quinone channel to accelerate the quinone/quinolone exchange. A number of conserved lipid and quinone binding sites have been identified, and we have revealed a new conformational change after the combination of quinone and RC, which may be suitable for photosystem II (PSII) RC of oxygenated phototrophic organisms. Our findings provide new insights into the kinetics of quinone/quinolone binding and exchange in the RC-LH1 core complex of purple phototrophic bacteria.
In order to facilitate a detailed study of the two complexes found in Rps. palustris, we isolate each RC-LH1 by biochemical methods. The protein W-deficient complex (hereinafter referred to as ΔpufW) was purified from the strain lacking the pufW gene (16), and only one RC-LH1 complex can be produced. The protein W-containing complex is produced by a strain. The protein W of this strain is modified with a 10x His tag at its C-terminus, so that the protein W-containing complex can be effectively combined with most lacking protein W by immobilizing metal. The complex is effectively separated (16) Affinity Chromatography (IMAC).
As shown in Figure 1, both complexes contain a three sub-unit RC (RC-L, RC-M and RC-H) surrounded by an LH1 antenna. The 2.80-A structure of the complex lacking protein-W shows 16 αβ heterodimers, forming a closed LH1 loop completely surrounding RC, hereinafter referred to as the RC-LH116 complex. The 2.65Å structure of the protein-W-containing complex has a 14-heterodimer LH1 interrupted by protein-W, hereinafter referred to as RC-LH114-W.
(A and B) Surface representation of the compound. (C and D) Bonded pigments expressed in rods. (E and F) The complexes observed from the cytoplasmic surface have the peptides and LH1 subunits represented in cartoons, and are numbered clockwise from the protein-W gap [consistent with Rba numbering. sphaeroides complex (13)]. For LH1-α, the color of the protein subunit is yellow; for LH1-β, the color of the protein subunit is blue; for protein-W, the protein is red; for RC-H, it is cyan; for RC-L, it is Orange; for RC-M, magenta. Cofactors are represented by rods, green represents BChl and BPh a molecules, purple represents carotenoids, and yellow represents UQ10 molecules. (G and H) Magnified view of the protein-W gap in the equivalent region of RC-LH114-W complex (G) and RC-LH116 complex (H). Cofactors are displayed in the form of space filling, chelated quinone is displayed in blue. The protein-W gap is highlighted by a blue dashed line in (G), and the small holes where quinone/quinolol diffuses on the LH116 ring are highlighted by a black dashed line in (H).
Figure 1 (A and B) shows the RC surrounded by open or closed arrays of LH1αβ heterodimers, each of which binds two BChl and one carotenoid (Figure 1, C and D). Previous studies have shown that Rps is the LH1 complex. In the biosynthetic pathway of spirulina xanthin, these species contain mixed populations of carotenoids (17). However, spiropyrroxanthin is the dominant carotenoid and its density is satisfactory. Therefore, we chose to model spiroxanthin at all LH1 binding sites. The alpha and beta polypeptides are single TMHs with short membrane outer regions (Figure 1, A, B, E, and F). Although the density of 17 residues at the C-terminus was not observed, the alpha polypeptide was cleaved from Met1 to Ala46 in both complexes. β polypeptide was reduced from Gly4 to Tyr52 in RC-LH116, and from Ser5 to Tyr52 in RC-LH114-W. No density of 3 or 4 N-terminal or 13 C-terminal residues was observed (Figure S1). Mass spectrometry analysis of the mixed RC-LH1 complex prepared from the wild-type strain showed that the missing region was the result of heterologous cleavage of these peptides (Figure S1 and S2). The N-terminal formylation of α-Met1 was also observed (f). The analysis showed that the α-peptide consists of residues fMet1 to Asp42/Ala46/Ala47/Ala50, and the β-peptide consists of residues Ser2 to Ala53, which is in good agreement with the low-temperature EM density map.
Coordination of α-His29 and β-His36 makes BChls face to face; each αβ heterodimer assembles with its neighbors to form an open loop (RC-LH114-W) or a closed loop (RC-LH116) around the RC The exciton coupled pigment array (Figure 1, C and D). Compared with the 877 nm band of RC-LH114-W, the 880 nm absorption red shift of RC-LH116 is 3 nm (Figure 2A). However, the circular dichroism spectrum is almost the same (Figure 2B), indicating that although there is a clear difference between open and closed loops, the local environment of BChls is very similar. The absorption redshift may be the result of reduced thermal motion and increased stability on the closed loop (18, 19), the change in pigment coupling caused by the closed loop (20, 21), or a combination of these two effects (11).
(A) Ultraviolet/visible/near-infrared absorption spectrum, the peaks of which are marked with their corresponding pigments and normalized to the BPh peak at 775 nm. (B) Circular dichroism spectrum normalized to BChl absorbance at 805 nm. (C and D) Selected ΔA spectra from the time-resolved absorption spectra of RC-LH114-W complex (C) and RC-LH116 complex (D). For better comparability, all spectra are normalized to ∆A of −A at 0.2 ps. (E) The rate of cytochrome c2 oxidation after irradiation in the presence of various concentrations of UQ2 (see Figure S8 for raw data). (F) In cells grown under low, medium or high intensity light (10, 30 or 300μMm-2 s-1, respectively), the protein W and RC-L subunits in the purified complex and the separated membrane ratio. Determine the protein level by SDS-polyacrylamide gel electrophoresis and immunoassay (see Figure S9 for raw data). Determine the ratio relative to the purified RC-LH114-W complex. The stoichiometric ratio of RC-L to protein-W of the complex is 1:1.
The BChls at position 1 in the deformed αβ14 loop of RC-LH114-W (Figure 1, A, C, and E) are closer to the RC primary donor (P) by 6.8Å than the equivalent BChls in RC-LH116 (Figure. 1, B, D, and F, and Figure S3); however, the transient absorption kinetics of the two complexes show that for RC-LH114-W and RC-LH116, the excitation energy transfer time constants from LH1 to RC are 40 ±4 and 44±3 ps (Figure 2). , C and D, Figure S4 and Table S2). There is also no significant difference in electronic transfer within RC (Figure S5 and related supplementary text). We suspect that the close correspondence of the energy transfer time between LH1 and RC-P is due to the similar distance, angle and potential energy of most BChl in the two LH1 loops. It seems that exploring the LH1 energy pattern to reach the minimum distance is not faster than direct energy transfer from suboptimal sites to RC. The open-loop LH1 loop in RC-LH114-W may also undergo insignificant thermal motion under low temperature conditions for structural analysis, and there is a longer αβ14 ring conformation at room temperature from the pigmentation distance of βBChls at the position of RC 1.
The RC-LH116 complex contains 32 BChls and 16 carotenoids, and its overall arrangement is the same as that obtained from Thermochromatium (Tch.) pidpidum [Protein Data Bank (PDB) ID 5Y5S] (9), Thiorhodovibrio (Trv.) 970 strain ( PDB ID 7C9R) (12) and green algae (Blc.viridis) (PDB ID 6ET5) (10). After alignment, only small deviations in the positions of αβ heterodimers were observed, especially 1-5, 15, and 16 (Figure S6). The presence of protein-W has a significant impact on the structure of LH1. Its three TMHs are connected by short loops, with the N-terminal on the lumen side of the complex and the C-terminal on the cytoplasmic side (Figures 1A and 3, A to D). Protein-W is largely hydrophobic (Figure 3B), and TMH2 and TMH3 interact with LH1αβ-14 to form a transmembrane surface (Figure 3, B and E to G). The interface is mainly composed of Phe, Leu and Val residues in the transmembrane region. These residues are stacked with hydrophobic amino acids and αβ-14 pigments. Some polar residues also contribute to the interaction, including the hydrogen bond between W-Thr68 and β-Trp42 on the surface of the complex cavity (Figure 3, F and G). On the surface of the cytoplasm, Gln34 is adjacent to the keto group of αβ-14 carotenoids. In addition, the n-dodecyl β-d-maltoside (β-DDM) molecule was resolved, and its hydrophobic tail extended to the interface between protein-W and αβ-14, and the lipid tail may be located in the body. We also noticed that the C-terminal resolution regions of protein W and RCH are very close, but not within the scope of forming specific interactions (Figure 1, A and E). However, there may be interactions in the unresolved C-terminal amino acids of these two proteins, which may provide a mechanism for the recruitment of protein-W during the assembly of the RC-LH114-W complex.
(A) Protein-W, which faces the interface with LH1αβ14 in cartoon form, has a rod-shaped side chain (red), displayed in a part of the electrostatic potential diagram (transparent gray surface with a contour level of 0.13). (B) Protein-W represented by a hydrophobic colored surface. Polar and charged areas are displayed in cyan, hydrophobic areas are displayed in white, and strongly hydrophobic areas are displayed in orange. (C and D) Protein-W represented in cartoon, its orientation is the same as in (A) (C), and rotated by 180° (D). According to the position in the sequence, the distinguishable residues adopt a rainbow color scheme, where the N-terminal is blue and the C-terminal is red. (E) Protein-W in the same view as in (A), and the residues at the interface of protein-W:LH1 are represented by rods with attached marks. (F) Protein-W is rotated 90° relative to (E) and LH1αβ14 in the cartoon representation, and relative to the interface residues in the bar representation. The overhanging residues from the beta polypeptide are labeled. The cofactor is shown as a bar matching the color of Figure 1, the decomposed β-DDM is shown in gray, and the oxygen is shown in red. (G) The view in (F) is rotated 180°, with the prominent residues of the labeled alpha polypeptide.
Protein-W replaces an αβ heterodimer (the 15th in Figure 1F), thereby preventing loop closure and tilting the first three αβ heterodimers. It was observed that the maximum inclination angle of the first αβ-1 heterodimer relative to the film normal was 25° to 29° (Figure 1, A and E), which was formed by the 2° to 8° inclination of αβ-1 in RC A sharp contrast-LH116 (Figure 1, B and F). The second and third heterodimers are inclined at 12° to 22° and 5° to 10°, respectively. Due to the steric hindrance of RC, the tilt of αβ-1 does not include the second pair of αβ (which corresponds to the 16th αβ in Figure 1F), thus forming a clear gap in the LH1 ring (Figure 1, A and E) . Due to the lack of two αβ heterodimers, accompanied by the loss of four BChl and two carotenoids, none of the carotenoids bind to the twisted αβ-1 subunit, resulting in a LH114-W ring containing 13 carotenoids Vegetarian and 28 BChls. The local resolution estimates of the two complexes in the αβ1 to 7 regions are lower than those of the rest of the LH1 loop, which may reflect the inherent plasticity of the LH1 subunit adjacent to the RC QB site (Figure 4) .
The pictures of RC-LH114-W (A and B) and RC-LH116 (C and D) are shown from the same top view/side view (A and B) (A and C) and cavity surface of Fig. 1. (B and D). The colored keys are shown on the right.
The only other characteristic core complex with a stoichiometric ratio of 1:14 is the Rhodococcus sphaeroides (Rba.) RC-LH1-PufX dimer (13). However, protein W and PufX have no obvious homology, and have a significant impact on their respective LH1 structures. PufX is a single TMH with an N-terminal cytoplasmic domain that interacts with the cytoplasmic side of the RC-H subunit (13) at a position corresponding to Rps. palustris LH116αβ-16. PufX creates a channel for the quinone/quinolone exchange between RC-LH1 and the cytochrome bcl complex and is present in all Rba. sphaeroides core complex (13). Although the monomer-monomer interface is in Rba. The sphaeroides RC-LH1-PufX dimer is located at the binding position of protein W in RC-LH114-W, and the gap induced by PufX and protein-W is at an equivalent position (Figure S7A). The gap in RC-LH114-W is also aligned with the hypothetical quinone channel (8) of Pseudomonas rosea LH1, which is formed by peptides not related to protein W or PufX (Figure S7B). In addition, the quinone channel in Blc. The emerald green LH1 formed by excluding one γ subunit (7) is located in a similar position (Figure S7C). Although mediated by different proteins, the appearance of these quinone/quinolol channels in a common position in the RC-LH1 complex seems to be an example of convergent evolution, indicating that the gap created by protein W may act as a quinone channel.
The gap in the LH114-W loop allows the formation of a continuous membrane region between the internal space of the RC-LH114-W complex and the bulk membrane (Figure 1G), rather than connecting the two domains through a protein pore as in proteins. The RC-LH116 complex is similar to a closed Tch. Needle-like complex (22) (Figure 1H). Since the diffusion of quinone through the membrane is faster than the diffusion through the narrow protein channel, the open LH114-W loop can allow faster RC turnover than the closed LH116 loop, and the diffusion of quinone into the RC may be more restricted. In order to test whether protein W affects the conversion of quinones through RC, we performed a cytochrome oxidation assay on a certain concentration of ubiquinone 2 (UQ2) (an analogue of natural UQ10 with a shorter isoprene tail) (Figure 2E). Although the presence of chelated quinone hinders the accurate determination of apparent Michaelis constant (RC-LH114-W and RC-LH116 are suitable for 0.2±0.1μM and 0.5±0.2μM, respectively), the maximum rate of RC-LH114-W ( 4.6±0.2 e-RC-1 s-1) is 28±5% larger than RC-LH116 (3.6±0.1 e-RC-1 s-1).
We initially estimated that protein-W is present in about 10% of the core complex (16); here, the occupancy rates of low-light, medium-light, and high-light growth cells are 15±0.6%, 11±1% and 0.9±0.5, respectively % (Figure 2F). Quantitative comparison of mass spectrometry showed that the addition of histidine tag did not reduce the relative abundance of protein-W compared to wild-type strains (P = 0.59), so these levels are not an artifact of modified protein-W (Figure S10). However, this low occupancy of protein-W in the RC-LH1 complex may allow some RCs to flip at an accelerated rate, thereby mitigating the slower quinone/quinolone exchange in the RC-LH116 complex. We noticed that the high light occupancy rate is inconsistent with the recent transcriptomics data, which indicates that pufW gene expression increases under strong light (Figure S11) (23). The difference between pufW transcription and protein-W incorporation into the RC-LH1 complex is confusing and may reflect the complex regulation of the protein.
In RC-LH114-W, 6 cardiolipin (CDL), 7 phosphatidylcholine (POPC), 1 phosphatidylglycerol (POPG) and 29 β-DDM molecules are allocated and modeled in it 6 CDLs, 24 POPCs, 2 POPGs and 12 βDDMs. RC-LH116 (Figure 5, A and B). In these two structures, CDL is almost located on the cytoplasmic side of the complex, while POPC, POPG and β-DDM are mostly located on the luminal side. Two lipid and detergent molecules were isolated in the αβ-1 to αβ-6 region of the RC-LH114-W complex (Figure 5A), and five were isolated in the equivalent region of RC-LH116 (Figure 5B ). More lipids were found on the other side of the complex, mainly CDL, accumulated between RC and αβ-7 to αβ-13 (Figure 5, A and B). Other structurally resolved lipids and detergents are located outside the LH1 ring, and well-resolved acyl chains extend between LH1 subunits, tentatively designated as β-DDM in RC-LH114-W, and defined as β-DDM in RC A mixture of β-DDM and POPC-LH116. The similar positions of chelating lipids and detergents in our structure indicate that they are physiologically relevant binding sites (Figure S12A). The positions of equivalent molecules in Tch also have good consistency. Gentle and Trv. Strain 970 RC-LH1s (Figure S12, B to E) (9, 12) and the hydrogen-bonding residues of the lipid head group showed fairly good conservation in the sequence alignment (Figure S13), indicating that Conserved CDL that binds to RC (24), these sites may be conserved in the RC-LH1 complex.
(A and B) RC-LH114-W (A) and RC-LH116 (B) peptides are represented by cartoons, and the pigments are represented by rods, using the color scheme in Figure 1. Lipids are shown in red, and detergents are shown in gray. UQ bound to RC QA and QB sites is yellow, while isolated UQ is blue. (C and D) The same views as (A) and (B), with lipids omitted. (E to G) Enlarged view of Q1(E), Q2(F) and Q3(G) from RC-LH116, with side chains that influence each other. The hydrogen bonds are shown as black dashed lines.
In RC-LH116, both RC QA and QB UQ, which participate in electron transfer in the charge separation process, are decomposed in their binding sites. However, in RC-LH114-W, QB quinone has not been resolved and will be discussed in detail below. In addition to QA and QB quinones, two chelated UQ molecules (located between the RC and LH1 rings) are allocated in the RC-LH114-W structure according to their well-resolved head groups (located in Q1 and Q2, respectively). space). Figure 5C). Two isoprene units are assigned to Q1, and the density map resolves the complete 10 isoprene tails of Q2. In the structure of RC-LH116, three chelated UQ10 molecules (Q1 to Q3, Figure 5D) were resolved, and all molecules have a clear density throughout the tail (Figure 5, D to G). In the two structures, the positions of the quinone head groups of Q1 and Q2 have excellent consistency (Figure S12F), and they only interact with RC. Q1 is located at the entrance of the W gap of RC-LH114-W (Figure 1G and 5, C, D and E), and Q2 is located near the QB binding site (Figure 5, C, D) and F). The conserved L-Trp143 and L-Trp269 residues are very close to Q1 and Q2 and provide potential π-stacking interactions (Figure 5, E and F, and Figure S12). L-Gln88, 3.0 Å from the distal oxygen of Q1, provides a strong hydrogen bond (Figure 5E); this residue is conserved in all RCs except the most distant relationship (Figure S13). L-Ser91 is conservatively substituted for Thr in most other RCs (Figure S13), is 3.8 Angstroms from the methyl oxygen of Q1, and may provide weak hydrogen bonds (Figure 5E). Q3 does not appear to have a specific interaction, but is located in the hydrophobic region between the RC-M subunit and the LH1-α subunit 5 to 6 (Figure 5, D and G). Q1, Q2 and Q3 or nearby chelated quinones have also been resolved in Tch. Gentle, Trv. Strain 970 and Blc. The iris structure (9, 10, 12) points to a conserved auxiliary quinone binding site in the RC-LH1 complex (Figure S12G). The five decomposed UQs in RC-LH116 are in good agreement with the 5.8±0.7 of each complex determined by high performance liquid chromatography (HPLC), while the three decomposed UQs in RC-LH114-W are lower than The measured value of 6.2±0.3 (Fig. S14) indicates that there are unresolved UQ molecules in the structure.
The pseudo-symmetric L and M polypeptides each contain five TMHs and form a heterodimer that combines one BChl dimer, two BChl monomers, two bacteriophage (BPh) monomers, and one non- Heme iron and one or two UQ10 molecules. Through the presence of hydrogen bonds on the terminal ketone group and its known accumulation in Rps, carotenoids are incorporated in the M-subunit, which is named cis-3,4-dehydroorhodopin. Species (25). The outer membrane domain of RC-H is anchored to the membrane by a single TMH. The overall RC structure is similar to the three subunit RC of related species (such as Rba). sphaeroides (PDB ID: 3I4D). The macrocycles of BChl and BPh, the carotenoid backbone and non-heme iron overlap within the resolution range of these structures, as do the UQ10 head group at the QA site and the QB quinone at RC-LH116 (Figure S15).
The availability of two RC structures with different QB site occupancy rates provides a new opportunity to examine the consistent conformational changes accompanying QB quinone binding. In the RC-LH116 complex, QB quinone is located in the fully bound “proximal” position (26), but the separation of RC-LH114-W does not have QB quinone. There is no QB quinone in RC-LH114-W, which is surprising because the complex is active, more so than the RC-LH116 complex with structurally resolved QB quinone. Although the two LH1 rings chelate about six quinones, five are structurally resolved in the closed RC-LH116 ring, while only three are structurally limited in the open RC-LH114-W ring . This increased structural disorder may reflect the faster replacement of RC-LH114-W QB sites, faster quinone kinetics in the complex, and increased likelihood of crossing the LH1 loop. We suggest that the lack of UQ in the RC QB site of RC-LH114-W may be the result of a more complex and more active complex, and the QB site of RC-LH114-W has been immediately frozen in UQ turnover The specific stage (the entrance to the QB site has been closed) reflects the conformation of this activity.
Without QB, the accompanying rotation of L-Phe217 to a position that is incompatible with UQ10 binding, because it will cause a spatial collision with the first isoprene unit of the tail (Figure 6A). In addition, the obvious main conformational changes are obvious, especially the helix de (short helix in the loop between TMH D and E) where L-Phe217 is shifted to the QB binding pocket and the rotation of L-Tyr223 (Figure 6A ) To break the hydrogen bond with the M-Asp45 framework and close the entrance of the QB binding site (Figure 6B). Helix de pivots at its base, the Cα of L-Ser209 is shifted by 0.33Å, while the L-Val221Cα is shifted by 3.52Å. There are no observable changes in TMH D and E, which are superimposable in both structures (Figure 6A). As far as we know, this is the first structure in the natural RC that closes the QB site. A comparison with the complete (QB-bound) structure shows that before the quinone is reduced, a conformational change is required to make it enter the quinone. L-Phe217 rotates to form a π-stacking interaction with the quinone head group, and the helix shifts outward, allowing the skeleton of L-Gly222 and the side chain of L-Tyr223 to form a hydrogen bond network with a stable hydrogen bond structure (Figure 6, A and C).
(A) Overlapping cartoon of hologram (L chain, orange/M chain, magenta) and apo (gray) structure, in which the key residues are displayed in the form of a rod-like representation. UQ10 is represented by a yellow bar. The dotted line indicates the hydrogen bonds formed in the entire structure. (B and C) The surface representation of the apolipoprotein and the whole ring structure, highlighting the side chain oxygen of L-Phe217 in blue and L-Tyr223 in red, respectively. The L subunit is orange; the M and H subunits are not colored. (D and E) Apolipoprotein (D) and whole (E) RC QB sites [color by (A) respectively] and Thermophilus thermophilus PSII (green, blue with plastic quinone; PDB ID: 3WU2) Align (58).
Unexpectedly, although several structures of QB-deficient RCs without LH1 are available, the conformational changes observed in this study have not been previously reported. These include the QB depletion structure from Blc. viridis (PDB ID: 3PRC) (27), Tch. tepidum (PDB ID: 1EYS) (28) and Rba. sphaeroides (PDB ID: 1OGV) (29), all of which are almost the same as their overall QB structure. Close inspection of 3PRC revealed that LDAO (Lauryl Dimethyl Amine Oxide) detergent molecules bind at the entrance of the QB position, which may prevent rearrangement into a closed conformation. Although LDAO does not decompose at the same position in 1EYS or 1OGV, these RCs are prepared using the same detergent and therefore may produce the same effect. The crystal structure of Rba. Sphaeroides RC co-crystallized with cytochrome c2 (PDB ID: 1L9B) also seems to have a closed QB site. However, in this case, the N-terminal region of the RC-M polypeptide (interacting with the QB binding site through the H bond of the Tyr residue on the Q helix) adopts an unnatural conformation, and the QB conformational change is not further explored (30 ). What is reassuring is that we have not seen this kind of deformation of the M polypeptide in the RC-LH114-W structure, which is almost the same as the N-terminal region of RC-LH116 RC. It should also be noted that after the eradication of the detergent-based LH1 antenna, the apolipoprotein RCs in the PDB were resolved, which eliminated the internal quinone pools and lipids in the gap between the RC and the inner surface of the surrounding LH1 ring (31, 32). RC remains functional because it retains all cofactors, except for the decomposable QB quinone, which is less stable and is often lost during the preparation process (33). In addition, it is known that the removal of LH1 and natural cyclic lipids from RC can have an impact on functions, such as the shortened lifespan of the charge-separated P+QB-state (31, 34, 35). Therefore, we speculate that the existence of the local LH1 ring surrounding the RC may maintain the “closed” QB site, thereby preserving the local environment near the QB.
Although apolipoprotein (without QB quinone) and the complete structure represent only two snapshots of the turnover of the QB site, rather than a series of events, there are indications that the binding can be gated to prevent rebinding by hydroquinone To inhibit substrate inhibition. The interaction of quinolol and quinone near the QB site of apolipoprotein may be different, which leads to its rejection by RC. It has long been proposed that conformational changes play a role in the binding and reduction of quinones. The ability of frozen RCs to reduce quinones after dark adaptation is impaired (36); X-ray crystallography shows that this damage is due to QB quinones being trapped in a “distal” conformation about 4.5 Å from the active proximal position (26) , 37). We suggest that this distal binding conformation is a snapshot of the intermediate state between apolipoprotein and the full ring structure, which follows the initial interaction with quinone and the opening of the QB site.
The type II RC found in the PSII complex of certain phototrophic bacteria and cyanobacteria, algae and plants has structural and functional conservation (38). The structural alignment shown in Figure 6 (D and E) emphasizes the similarity between PSII RCs and the QB site of the bacterial RC complex. This comparison has long been a model for studying the closely related systems of quinone binding and reduction. Previous publications suggested that conformational changes are accompanied by PSII reduction of quinones (39, 40). Therefore, considering the evolutionary conservation of RC, this previously unobserved binding mechanism may also be applicable to the QB site of PSII RC in oxygenated phototrophic plants.
Rps ΔpufW (unlabeled pufW deletion) and PufW-His (C-terminal 10x His-tagged protein-W expressed from the natural pufW locus) strains. palustris CGA009 was described in our previous work (16). These strains and the isogenic wild-type parent were recovered from the freezer by streaking a small number of cells on PYE (each 5 g liter -1) (stored in LB at -80 °C, containing 50% ( w/v) glycerol) protein, yeast extract and succinate) agar [1.5% (w/v)] plate. The plate was incubated overnight in the dark at room temperature under anaerobic conditions, and then illuminated with white light (~50 μmolm-2 s-1) provided by OSRAM 116-W halogen bulbs (RS Components, UK) for 3 to 5 days until a single colony appeared. A single colony was used to inoculate 10 ml of M22+ medium (41) supplemented with 0.1% (w/v) casamino acids (hereinafter referred to as M22). The culture was grown under low oxygen conditions in the dark at 34°C with shaking at 180 rpm for 48 hours, and then 70 ml of the culture was inoculated under the same conditions for 24 hours. A semi-aerobic culture with a volume of 1 ml is used to inoculate 30 ml of M22 medium in a 30 ml universal screw-top transparent glass bottle and irradiated with agitation (~50μmolm-2 s-1) for 48 hours by sterile magnetic force Stirring rod. Then 30 ml of the culture was inoculated with about 1 liter of culture under the same conditions, which was then used to inoculate about 9 liters of culture illuminated at ~200 μmolm-2 s-1 for 72 hours. The cells were harvested by centrifugation at 7132 RCF for 30 minutes, resuspended in ~10 ml of 20 mM tris-HCl (pH 8.0), and stored at -20°C until needed.
After thawing, add some crystals of deoxyribonuclease I (Merck, UK), lysozyme (Merck, UK) and two Roche holoenzyme protease inhibitor tablets (Merck, UK) to the resuspended cells. In a 20,000 psi French pressure cell (Aminco, USA), the cells were disrupted 8 to 12 times. After removing unbroken cells and insoluble debris by centrifugation at 18,500 RCF for 15 minutes at 4°C, the membrane was precipitated from the pigmented lysate by centrifugation at 113,000 RCF for 2 hours at 43,000°C. Discard the soluble fraction and resuspend the colored membrane in 100 to 200 ml of 20 mM tris-HCl (pH 8.0) and homogenize until there are no visible aggregates. The suspended membrane was incubated in 20 mM tris-HCl (pH 8.0) (Anatrace, USA) containing 2% (w/v) β-DDM for 1 hour in the dark at 4°C with gentle stirring. Then centrifuge at 70°C to dissolve 150,000 RCF at 4°C for 1 hour to remove residual insolubles.
The solubilizing membrane from the ΔpufW strain was applied to a 50 ml DEAE Sepharose ion exchange column with three column volumes (CV) of binding buffer [20 mM tris-HCl (pH 8.0) containing 0.03% (w / v) β-DDM]. Wash the column with two CV binding buffers, and then wash the column with two binding buffers containing 50 mM NaCl. The RC-LH116 complex was eluted with a linear gradient of 150 to 300 mM NaCl (in binding buffer) on 1.75 CV, and the remaining binding complex was eluted with a binding buffer containing 300 mM NaCl on 0.5 CV. Collect the absorption spectrum between 250 and 1000 nm, keep the fraction with absorbance ratio (A880/A280) greater than 1 at 880 to 280 nm, dilute it twice in the binding buffer, and use the same procedure again on the DEAE column On purification. Dilute the fractions with A880/A280 ratios higher than 1.7 and A880/A805 ratios higher than 3.0, perform the third round of ion exchange, and retain fractions with A880/A280 ratios higher than 2.2 and A880/A805 ratios higher than 5.0. The partially purified complex was concentrated to ~2 ml in an Amicon 100,000 molecular weight cut-off (MWCO) centrifugal filter (Merck, UK), and loaded on a Superdex 200 16/600 size exclusion column (GE Healthcare, U.S.) containing 200 mM NaCl buffer, and then eluted in the same buffer at 1.5 CV. Collect the absorption spectra of the size exclusion fraction, and concentrate the absorption spectra with A880/A280 ratios over 2.4 and A880/A805 ratios over 5.8 to 100 A880, and immediately use them for cryo-TEM grid preparation or storage Keep at -80°C until needed.
The solubilizing membrane from PufW-His strain was applied to a 20 ml HisPrep FF Ni-NTA Sepharose column (20 mM tris-HCl (pH 8.0) containing 200 mM NaCl and 0.03% (w/w)) in IMAC buffer ( GE Healthcare). v) β-DDM]. The column was washed with five CVs of IMAC buffer, and then with five CVs of IMAC buffer containing 10 mM histidine. The core complex was eluted from the column with five IMAC buffers containing 100 mM histidine. The fraction containing the RC-LH114-W complex is concentrated to ~10 ml in a stirred tank equipped with an Amicon 100,000 MWCO filter (Merck, UK), diluted 20 times with binding buffer, and then added to 25 ml In the DEAE Sepharose column, four CVs bound to the buffer are used in advance. Wash the column with four CV binding buffers, then elute the complex on eight CVs on a linear gradient of 0 to 100 mM NaCl (in binding buffer), and the remaining four CVs containing 100 mM binding buffer The residual complexes eluted on the sodium chloride combined with the A880/A280 ratio higher than 2.4 and the A880/A805 ratio higher than 4.6 fractions were concentrated to ~2 ml in an Amicon 100,000 MWCO centrifugal filter, and filled with 1.5 CV IMAC in advance Buffer equilibrated Superdex 200 16/600 size exclusion column, and then eluted in the same buffer over 1.5 CV. Collect the absorption spectra of the size-exclusion fractions and concentrate the absorption spectra with A880/A280 ratios over 2.1 and A880/A805 ratios over 4.6 to 100 A880, which are immediately used for frozen TEM grid preparation or stored at -80°C until need.
A Leica EM GP immersion freezer was used to prepare low temperature TEM grids. The complex was diluted in IMAC buffer to A880 of 50, and then 5μl was loaded onto a newly glow-discharged QUANTIFOIL 1.2/1.3 carbon-coated copper mesh (Agar Scientific, UK). Incubate the grid at 20°C and 60% relative humidity for 30 s, then blot it dry for 3 s, and then quench it in liquid ethane at -176°C.
The data of the RC-LH114-W complex was recorded on the eBIC (Electronic Bioimaging Center) (British Diamond Light Source) with a Titan Krios microscope, which works at an accelerating voltage of 300kV, with a nominal magnification of 130,000× and an energy of- Choose a gap of 20 eV. A Gatan 968 GIF Quantum with K2 peak detector was used to record images in counting mode to collect data. The calibrated pixel size is 1.048Å, and the dose rate is 3.83 e-Å-2s-1. Collected the movie in 11 seconds and divided it into 40 parts. Use the carbon-coated area to refocus the microscope, and then collect three movies per hole. In total, 3130 movies were collected, with defocus values ​​between -1 and -3μm.
The data for the RC-LH116 complex was collected using the same microscope at the Asterbury Biostructure Laboratory (University of Leeds, UK). The data was collected in counting mode with a magnification of 130 k, and the pixel size was calibrated to 1.065 Å with a dose of 4.6 e-Å-2s-1. The movie was recorded in 12 seconds and divided into 48 parts. In total, 3359 films were collected, with defocus values ​​between -1 and -3μm.
All data processing is performed in the Relion 3.0 pipeline (42). Use Motioncorr 2 (43) to correct the beam motion by dose weighting, and then use CTFFIND 4.1 (44) to determine the CTF (contrast transfer function) parameter. Typical photomicrographs after these initial processing stages are shown in Figure 2. S16. The automatic selection template is generated by manually selecting about 250 pixels of 1000 particles in a 250-pixel frame and no reference two-dimensional (2D) classification, thereby rejecting those classifications that meet sample contamination or have no discernible characteristics. Then, automatic selection was performed on all the microphotographs, and the RC-LH114-W was 849,359 particles, and the RC-LH116 complex was 476,547 particles. All selected particles have undergone two rounds of non-reference 2D classification, and after each run, the particles that meet the carbon area, sample contamination, no obvious features or strongly overlapping particles are rejected, resulting in 772,033 (90.9%) and 359,678 (75.5%) ) Particles are used for 3D classification of RC-LH114-W and RC-LH116 respectively. The initial 3D reference model was generated using the stochastic gradient descent method. Using the initial model as a reference, the selected particles are classified into four categories in 3D. Using the model in this category as a reference, perform 3D refining on the particles in the largest category, then use the initial 15Å low-pass filter to cover the solvent area, add 6 pixels of soft edges, and post-process the pixels to correct the Gatan K2 peak Modulation transfer function of the top detector. For the RC-LH114-W dataset, this initial model was modified by removing the strong density at the edges of the mask (disconnected from the core complex density in UCSF Chimera). The resulting models (the resolutions of RC-LH114-W and RC-LH116 are 3.91 and 4.16 Å, respectively) are used as a reference for the second round of 3D classification. The particles used are grouped into the initial 3D class and do not contain strong correlation with the neighborhood. Overlap or lack of obvious structural features. After the second round of 3D classification, the category with the highest resolution was selected [For RC-LH114-W, one category is 377,703 particles (44.5%), for RC-LH116, there are two categories, totaling 260,752 particles (54.7%) , Where they are the same only when aligned after the initial rotation with a small difference]. The selected particles are re-extracted in a 400-pixel box and refined by 3D refining. The solvent mask is generated using the initial 15Å low-pass filter, 3 pixel map expansion and 3 pixel soft mask. Using per-particle CTF refinement, per-particle motion correction and the second round of per-particle CTF refinement, 3D refinement, solvent masking and post-processing are performed after each step to further refine the resulting texture. Using the FSC (Fourier Shell Correlation Coefficient) cut-off value of 0.143, the resolutions of the final models of RC-LH114-W and RC-LH116 are 2.65 and 2.80Å, respectively. The FSC curve of the final model is shown in Figure 2. S17.
All protein sequences are downloaded from UniProtKB: LH1-β (PufB; UniProt ID: Q6N9L5); LH1-α (PufA; UniProtID: Q6N9L4); RC-L (PufL; UniProt ID: O83005); RC-M (PufM; UniProt ID: A0A4Z7); RC-H (PuhA; UniProt ID: A0A4Z9); Protein-W (PufW; UniProt ID: Q6N1K3). SWISS-MODEL (45) was used to construct a homology model of RC, which contains the protein sequences of RC-L, RC-M and RC-H and the crystal structure of Rba. sphaeroides RC was used as a template (PDB ID: 5LSE) (46). Use the “fit map” tool in UCSF Chimera to fit the generated model to the map (47), improve the protein structure, and cofactor [4×BChl a (monomer library residue name = BCL), 2×BPh a (BPH), one or two kinds of UQ10 (U10), one non-heme iron (Fe) and one 3,4-dihydrohexacarbonylcholine (QAK)] use Coot (48) to add. Since QAK is not available in the monomer library, it was parameterized using the eLBOW tool in PHENIX (49).
Next, the LH1 subunit was constructed. Initially, the automatic construction tool in PHENIX (49) was used to automatically construct part of the LH1 sequence using the map and the LH1-α and LH1-β protein sequences as input. Select the most complete LH1 subunit, extract it and load it into Coot, manually add the missing sequence in it, and manually refine the entire structure before adding two BCls a (BCL) and a spirilloxanthin (CRT) [according to the relevant Rps The density of the LH1 complex and the known carotenoid content. Species (17)]. Copy the complete LH1 subunit, and use the UCSF Chimera “Docking Map Tool” to dock in the adjacent non-model area of ​​LH1 density, and then refine it in Coot; repeat the process until all LH1 subunits have been modeled. For the RC-LH114-W structure, by extracting the unallocated density in the Coot, the protein is segmented from the remaining non-protein components in the USCF Chimera map and the Autobuild tool is used to establish the initial model, and the remaining subunits (protein-W) Modeling. In PHENIX (49). Add any missing sequences to the resulting model in Coot (48), and then manually refine the entire subunit. The remaining unallocated density fits the combination of lipids (PDB monomer library ID of CDL = CDL, POPC = 6PL and POPG = PGT), β-DDM detergent (LMT) and UQ10 molecules (U10). Use PHENIX optimization (49) and manual optimization in Coot (48) to perfect the complete initial model until the model statistics and the visual quality of the fit cannot be further improved. Finally, use LocScale (50) to sharpen the local map, and then perform several other cycles of modeling the unallocated density and automatic and manual optimization.
The respective peptides, cofactors and other lipids and quinones docked within their respective densities are shown in Figures 1 and 2. S18 to S23. The statistical information of the final model is shown in Table S1.
Unless otherwise specified, the UV/Vis/NIR absorption spectra were collected on a Cary60 spectrophotometer (Agilent, USA) at 1 nm intervals from 250 nm to 1000 nm and an integration time of 0.1s.
Dilute the sample in a quartz cuvette with a 2 mm path to A880 of 1, and collect the absorption spectrum between 400 and 1000 nm. The circular dichroic spectra were collected on a Jasco 810 spectropolarimeter (Jasco, Japan) at 1 nm intervals between 400 nm and 950 nm at a scan rate of 20 nm min-1.
The molar extinction coefficient is determined by diluting the core complex to an A880 of approximately 50. Dilute the 10μl volume in 990μl binding buffer or methanol, and collect the absorption spectrum immediately to minimize BChl degradation. The BChl content of each methanol sample was calculated by the extinction coefficient at 771 nm of 54.8 mM-1 cm-1, and the extinction coefficient was determined (51). Divide the measured BChl concentration by 32 (RC-LH114-W) or 36 (RC-LH116) to determine the core complex concentration, which is then used to determine the absorption spectrum of the same sample collected in the buffer Extinction coefficient. parallel. Three repeated measurements were taken for each sample, and the average absorbance of the BChl Qy maximum was used for calculation. The extinction coefficient of RC-LH114-W measured at 878 nm is 3280±140 mM-1 cm-1, while the extinction coefficient of RC-LH116 measured at 880 nm is 3800±30 mM-1 cm-1.
UQ10 was quantified according to the method in (52). In short, reverse phase HPLC (RP-HPLC) was performed using the Agilent 1200 HPLC system. Dissolve about 0.02 nmol of RC-LH116 or RC-LH114-W in 50μl of 50:50 methanol:chloroform containing 0.02% (w/v) ferric chloride, and inject the pre-equilibrated Beckman Coulter Ultrasphere ODS 4.6 mm Dissolve in 1 ml-1 min-1 at 40°C in HPLC solvent (80:20 methanol:2-propanol) on a ×25 cm column. Perform isocratic elution in an HPLC solvent to monitor absorbance at 275 nm (UQ10), 450 nm (carotenoids) and 780 nm (BChl) for 1 hour. The peak in the 275 nm chromatogram at 25.5 minutes was integrated, which did not contain any other detectable compounds. The integrated area is used to calculate the molar amount of UQ10 extracted with reference to the calibration curve calculated from the injection of pure standards from 0 to 5.8 nmol (Figure S14). Each sample was analyzed in three replicates, and the reported error corresponds to the SD of the average.
A solution containing the RC-LH1 complex with a maximum Qy absorption of 0.1 was prepared with 30 μM reduced horse heart cytochrome c2 (Merck, UK) and 0 to 50 μMUQ2 (Merck, UK). Three 1-ml samples were prepared at each UQ2 concentration and incubated overnight in the dark at 4°C to ensure complete adaptation to the dark before measurement. The solution was loaded into an OLIS RSM1000 modular spectrophotometer equipped with a 300 nm flame/500 line grating, 1.24 mm inlet, 0.12 mm middle and 0.6 mm outlet slits. A 600 nm long pass filter is placed at the entrance of the sample phototube and the reference photomultiplier tube to exclude excitation light. The absorbance was monitored at 550 nm with an integration time of 0.15 s. The excitation light is emitted from the 880 nm M880F2 LED (Light Emitting Diode) (Thorlabs Ltd., UK) through a fiber optic cable at 90% intensity through a DC2200 controller (Thorlabs Ltd., UK) and is emitted to the light source at an angle of 90° of. The measuring beam is opposed to the mirror to return any light that was not initially absorbed by the sample. Monitor the absorbance 10 s before the illuminance of 50 s. Then the absorbance was further monitored for 60 s in the dark to assess the extent to which quinolol spontaneously reduces cytochrome c23 + (see Figure S8 for raw data).
The data was processed by fitting a linear initial rate within 0.5 to 10 s (depending on the UQ2 concentration) and averaging the rates of all three samples at each UQ2 concentration. The RC-LH1 concentration calculated by the respective extinction coefficient was used to convert the rate into the catalytic efficiency, plotted in Origin Pro 2019 (OriginLab, USA), and fitted to the Michaelis-Menten model to determine the apparent Km and Kcat values.
For transient absorption measurements, the RC-LH1 sample was diluted to ~2μM in IMAC buffer containing 50 mM sodium ascorbate (Merck, USA) and 0.4 mM Terbutin (Merck, USA). Ascorbic acid is used as a sacrificial electron donor, and tert-butaclofen is used as a QB inhibitor to ensure that the main RC donor remains reduced (that is, not photooxidized) throughout the measurement process. Approximately 3 ml of sample is added to a custom rotating cell (about 0.1 m in diameter, 350 RPM) with a 2 mm optical path length to ensure that the sample in the laser path has enough time for dark adaptation between excitation pulses. Use ~100-fs laser pulses to amplify the Ti: Sapphire laser system (Spectra Physics, USA) to excite the sample at 880 nm at a repetition rate of 1 kHz (20 nJ for NIR or 100 nJ for Vis). Before collecting data, expose the sample to excitation light for about 30 minutes. Exposure will cause QA inactivation (possibly reducing QA once or twice). But please note that this process is reversible because after a long period of dark adaptation, RC will slowly return to QA activity. A Helios spectrometer (Ultrafast Systems, USA) was used to measure transient spectra with a delay time of -10 to 7000 ps​​. Use Surface Xplorer software (Ultrafast Systems, USA) to ungroup the data sets, then merge and standardize. Use the CarpetView software package (Light Conversion Ltd., Lithuania) to use the combined data set to obtain differential spectra related to decay, or use a function that convolves multiple exponents with the instrument response to fit the single-wavelength spectral evolution in Origin (OriginLab, USA).
As mentioned above (53), a photosynthetic film containing LH1 complex lacking both RC and peripheral LH2 antenna was prepared. The membrane was diluted in 20 mM tris (pH 8.0) and then loaded into a quartz cuvette with a 2 mm optical path. A 30nJ laser pulse was used to excite the sample at 540 nm with a delay time of -10 to 7000 ps. Process the data set as described for Rps. pal sample.
The membrane was pelleted by centrifugation at 150,000 RCF for 2 hours at 4°C, and then its absorbance at 880 nm was resuspended in 20 mM tris-HCl (pH 8.0) and 200 mM NaCl. Dissolve the membrane by slowly stirring in 2% (w/v) β-DDM for 1 hour in the dark at 4°C. The sample was diluted in 100 mM triethylammonium carbonate (pH 8.0) (TEAB; Merck, UK) to a protein concentration of 2.5 mg ml-1 (Bio-Rad analysis). Further processing was carried out from the previously published method (54), starting with the dilution of 50 μg protein into a total of 50 μl TEAB containing 1% (w/v) sodium laurate (Merck, UK). After sonication for 60 s, it was reduced with 5 mM tris(2-carboxyethyl)phosphine (Merck, UK) at 37°C for 30 minutes. For S-alkylation, incubate the sample with 10 mM methyl S-methylthiomethanesulfonate (Merck, UK) and add it from a 200 mM isopropanol stock solution for 10 minutes at room temperature. Proteolytic digestion was carried out by adding 2 μg trypsin/endoproteinase Lys-C mixture (Promega UK) and incubated at 37°C for 3 hours. The laurate surfactant was extracted by adding 50 μl ethyl acetate and 10 μl 10% (v/v) LC grade trifluoroacetic acid (TFA; Thermo Fisher Scientific, UK) and vortexing for 60 s. The phase separation was promoted by centrifugation at 15,700 RCF for 5 minutes. According to the manufacturer’s protocol, a C18 spin column (Thermo Fisher Scientific, UK) was used to carefully aspirate and desalt the lower phase containing the peptide. After drying by vacuum centrifugation, the sample was dissolved in 0.5% TFA and 3% acetonitrile, and 500 ng was analyzed by nanoflow RP chromatography coupled with mass spectrometry using the system parameters detailed previously.
Use MaxQuant v.1.5.3.30 (56) for protein identification and quantification to search for Rps. palustris proteome database (www.uniprot.org/proteomes/UP000001426). The mass spectrometry proteomics data has been deposited in the ProteomeXchange Alliance through the PRIDE partner repository (http://proteomecentral.proteomexchange.org) under the dataset identifier PXD020402.
For analysis by RPLC coupled with electrospray ionization mass spectrometry, the RC-LH1 complex was prepared from wild-type Rps. Using the previously published method (16), the protein concentration produced in palustris cells was 2 mg ml-1 in 20 mM Hepes (pH 7.8), 100 mM NaCl and 0.03% (w/v) β- (Bio-Rad analysis) ) DDM. According to the manufacturer’s protocol, use 2D purification kit (GE Healthcare, USA) to extract 10 μg protein by precipitation method, and dissolve the precipitate in 20 μl 60% (v / v) formic acid (FA), 20% (v / v) Acetonitrile and 20% (v/v) water. Five microliters were analyzed by RPLC (Dionex RSLC) coupled with mass spectrometry (Maxis UHR-TOF, Bruker). Use MabPac 1.2×100 mm column (Thermo Fisher Scientific, UK) for separation at 60°C and 100μlmin -1, with a gradient of 85% (v / v) solvent A [0.1% (v / v) FA and 0.02% (V/v) TFA aqueous solution] to 85%(v/v) solvent B [0.1%(v/v) FA and 0.02%(v/v) in 90%(v/v) acetonitrile TFA] Using standard electrospray ionization source and default parameters for more than 60 minutes, the mass spectrometer obtains 100 to 2750 m/z (mass-to-charge ratio). With the help of the ExPASy bioinformatics resource portal FindPept tool (https://web.expasy.org/findpept/), map the mass spectrum to the subunits of the complex.
The cells were grown for 72 hours under 100 ml NF-low (10μMm-2 s-1), medium (30μMm-2 s-1) or high (300μMm-2 s-1) light. M22 medium (M22 medium in which ammonium sulfate is omitted and sodium succinate is replaced by sodium acetate) in a 100 ml screw-top bottle (23). In five 30-s cycles, 0.1 micron glass beads were beaded at a volume ratio of 1:1 to lyse the cells and chilled on ice for 5 minutes. The insoluble matter, unbroken cells and glass beads were removed by centrifugation at 16,000 RCF for 10 minutes in a benchtop microcentrifuge. The membrane was separated in a Ti 70.1 rotor with 100,000 RCF in 20 mM tris-HCl (pH 8.0) with a 40/15% (w/w) sucrose gradient for 10 hours.
As described in our previous work, immunodetection of the His tag on PufW (16). In short, the purified core complex (11.8 nM) or the membrane containing the same concentration of RC (determined by oxidation subtracting the reduced difference spectrum and matching the load on the stained gel) in 2x SDS loading buffer (Merck, UK) diluted twice. The proteins were separated on a replica 12% bis-tris NuPage gel (Thermo Fisher Scientific, UK). A gel was stained with Coomassie Brilliant Blue (Bio-Rad, UK) to load and visualize the RC-L subunit. The protein on the second gel was transferred to methanol-activated polyvinylidene fluoride (PVDF) membrane (Thermo Fisher Scientific, UK) for immunoassay. The PVDF membrane was blocked in 50 mM tris-HCl (pH 7.6), 150 mM NaCl, 0.2% (v / v) Tween-20 and 5% (w / v) skimmed milk powder, and then incubated with the anti-His primary antibody (in Dilute the antibody buffer [50 mM tris-HCl (pH 7.6), 150 mM NaCl and 0.05% (v/v) Tween-20] in 1:1000 A190-114A, Bethyl Laboratories, USA) for 4 hours. After washing 3 times for 5 minutes in antibody buffer, the membrane was combined with horseradish peroxidase (Sigma-Aldrich, UK) anti-mouse secondary antibody (diluted 1:10,000 in antibody buffer) Incubate to allow detection (5 minutes after 3 washes in antibody buffer) using WESTAR ETA C 2.0 chemiluminescence substrate (Cyanagen, Italy) and Amersham Imager 600 (GE Healthcare, UK).
By drawing the intensity distribution of each stained gel or immunoassay lane, integrating the area under the peak and calculating the intensity ratio of RC-L (stained gel) and Protein-W (immunoassay), in ImageJ (57) Process the image. These ratios were converted to molar ratios by assuming that the ratio of RC-L to protein-W in the pure RC-LH114-W sample was 1:1 and normalizing the entire data set accordingly.
For supplementary materials for this article, please see http://advances.sciencemag.org/cgi/content/full/7/3/eabe2631/DC1
This is an open access article distributed under the terms of the Creative Commons Attribution License. The article allows unrestricted use, distribution, and reproduction in any medium under the condition that the original work is properly cited.
Note: We only ask you to provide your email address so that the person you recommend to the page knows that you want them to see the email and that it is not spam. We will not capture any email addresses.
This question is used to test whether you are a visitor and prevent automatic spam submission.
David JK Swainsbury, Park Qian, Philip J. Jackson, Kaitlyn M. Faries, Dariusz M. Niedzwiedzki, Elizabeth C. Martin, David A. Farmer, Lorna A. Malone, Rebecca F. Thompson, Neil A. Ranson, Daniel P Canniffe , Mark J. Dickman, Dewey Holten, Christine Kirmaier, Andrew Hitchcock, C. Neil Hunter
The high-resolution structure of the light trap 1 complex in the reaction center provides new insights into the quinone dynamics.
David JK Swainsbury, Park Qian, Philip J. Jackson, Kaitlyn M. Faries, Dariusz M. Niedzwiedzki, Elizabeth C. Martin, David A. Farmer, Lorna A. Malone, Rebecca F. Thompson, Neil A. Ranson, Daniel P Canniffe , Mark J. Dickman, Dewey Holten, Christine Kirmaier, Andrew Hitchcock, C. Neil Hunter
The high-resolution structure of the light trap 1 complex in the reaction center provides new insights into the quinone dynamics.
©2021 American Association for the Advancement of Science. all rights reserved. AAAS is a partner of HINARI, AGORA, OARE, CHORUS, CLOCKSS, CrossRef and COUNTER. ScienceAdvances ISSN 2375-2548.