Nanodiscs unravel the interaction between the SecYEG channel and its cytosolic partner SecA
2007; Springer Nature; Volume: 26; Issue: 8 Linguagem: Inglês
10.1038/sj.emboj.7601661
ISSN1460-2075
AutoresMériem Alami, Kush Dalal, Barbara Lelj‐Garolla, Stephen G. Sligar, Franck Duong,
Tópico(s)Advanced biosensing and bioanalysis techniques
ResumoArticle29 March 2007free access Nanodiscs unravel the interaction between the SecYEG channel and its cytosolic partner SecA Meriem Alami Meriem Alami Department of Biochemistry and Molecular Biology, Life Sciences Institute, Faculty of Medicine, University of British Columbia, British Columbia, Canada Search for more papers by this author Kush Dalal Kush Dalal Department of Biochemistry and Molecular Biology, Life Sciences Institute, Faculty of Medicine, University of British Columbia, British Columbia, Canada Search for more papers by this author Barbara Lelj-Garolla Barbara Lelj-Garolla Department of Biochemistry and Molecular Biology, Life Sciences Institute, Faculty of Medicine, University of British Columbia, British Columbia, Canada Search for more papers by this author Stephen G Sligar Stephen G Sligar Department of Biochemistry, University of Illinois at Urbana-Champaign, Urbana, IL, USA Search for more papers by this author Franck Duong Corresponding Author Franck Duong Department of Biochemistry and Molecular Biology, Life Sciences Institute, Faculty of Medicine, University of British Columbia, British Columbia, Canada Search for more papers by this author Meriem Alami Meriem Alami Department of Biochemistry and Molecular Biology, Life Sciences Institute, Faculty of Medicine, University of British Columbia, British Columbia, Canada Search for more papers by this author Kush Dalal Kush Dalal Department of Biochemistry and Molecular Biology, Life Sciences Institute, Faculty of Medicine, University of British Columbia, British Columbia, Canada Search for more papers by this author Barbara Lelj-Garolla Barbara Lelj-Garolla Department of Biochemistry and Molecular Biology, Life Sciences Institute, Faculty of Medicine, University of British Columbia, British Columbia, Canada Search for more papers by this author Stephen G Sligar Stephen G Sligar Department of Biochemistry, University of Illinois at Urbana-Champaign, Urbana, IL, USA Search for more papers by this author Franck Duong Corresponding Author Franck Duong Department of Biochemistry and Molecular Biology, Life Sciences Institute, Faculty of Medicine, University of British Columbia, British Columbia, Canada Search for more papers by this author Author Information Meriem Alami1, Kush Dalal1, Barbara Lelj-Garolla1, Stephen G Sligar2 and Franck Duong 1 1Department of Biochemistry and Molecular Biology, Life Sciences Institute, Faculty of Medicine, University of British Columbia, British Columbia, Canada 2Department of Biochemistry, University of Illinois at Urbana-Champaign, Urbana, IL, USA *Corresponding author. Department of Biochemistry and Molecular Biology, Life Sciences Institute, University of British Columbia, 2350 Health Sciences Mall, Vancouver, British Columbia, Canada V6T 1Z3. Tel.: +1 604 822 5975; Fax: +1 604 822 5227; E-mail: [email protected] The EMBO Journal (2007)26:1995-2004https://doi.org/10.1038/sj.emboj.7601661 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The translocon is a membrane-embedded protein assembly that catalyzes protein movement across membranes. The core translocon, the SecYEG complex, forms oligomers, but the protein-conducting channel is at the center of the monomer. Defining the properties of the SecYEG protomer is thus crucial to understand the underlying function of oligomerization. We report here the reconstitution of a single SecYEG complex into nano-scale lipid bilayers, termed Nanodiscs. These water-soluble particles allow one to probe the interactions of the SecYEG complex with its cytosolic partner, the SecA dimer, in a membrane-like environment. The results show that the SecYEG complex triggers dissociation of the SecA dimer, associates only with the SecA monomer and suffices to (pre)-activate the SecA ATPase. Acidic lipids surrounding the SecYEG complex also contribute to the binding affinity and activation of SecA, whereas mutations in the largest cytosolic loop of the SecY subunit, known to abolish the translocation reaction, disrupt both the binding and activation of SecA. Altogether, the results define the fundamental contribution of the SecYEG protomer in the translocation subreactions and illustrate the power of nanoscale lipid bilayers in analyzing the dynamics occurring at the membrane. Introduction The Sec translocon is a membrane protein complex that cooperates with cytosolic partners to drive polypeptide substrate translocation into and across membranes. The translocon may be comprised of multiple components, but the core is made of a conserved heterotrimer known as SecYEG in bacteria, SecYEβ in archaea and Sec61p in eukaryotes (Cao and Saier, 2003). The remarkable fact that the translocon forms a protein-conducting channel is now rationalized through structural analysis of the core complex: cryo-electron microscopy and high-resolution crystal structures recently unraveled the location of the protein channel path at the center of the SecY subunit (Van den Berg et al, 2004; Mitra et al, 2005). Three other key structural elements were identified following structural determinations: the ‘plug’ domain formed by a small helix blocking the channel on its periplasmic side; the ‘pore ring’ made by a few residues forming a constriction point at the center of the channel; and the ‘lateral gate’ created by the juxtaposition of two SecY transmembrane segments forming an opening toward the lipid bilayer. Recent biochemical experiments along with computer-based simulations support this model, including the widening of the pore and the displacement of the plug away from the center of the channel by the incoming polypeptide substrate (Cannon et al, 2005; Tam et al, 2005; Gumbart and Schulten, 2006; Tian and Andricioaei, 2006; Maillard et al, 2007). Despite this tremendous progress, there are major problems that still remain in the field. One of the major questions, complicated by the observation that the channel is contained within a single SecY subunit, is that the SecYEG complex spontaneously forms oligomers of two or more copies, both in the membrane or in detergent solution (Manting et al, 2000; Collinson et al, 2001; Bessonneau et al, 2002; Mori et al, 2003; for review see Rusch and Kendall, 2007). One original hypothesis assumed that the channel was formed at the center of these oligomers, but experimentation now supports the structural prediction that a single SecYEG complex forms the translocation pathway (Duong, 2003; Cannon et al, 2005). Nevertheless, additional evidence indicates that the SecYEG dimer and also probably the tetramer are essential and active oligomeric assemblies. Cryo-electron microscopy of a translocon engaged with a ribosome-nascent chain complex shows electron density representative of a SecYEG dimer (Mitra et al, 2005). Similarly, at the endoplasmic reticulum membrane, the active channel bound to ribosomes consists of four copies of Sec61p (Beckmann et al, 2001; Ménétret et al, 2005). However, the significance of SecYEG complex oligomerization remains puzzling. Defining the underlying function of monomer and oligomers in the translocation subreaction is thus crucial for the mechanistic understanding of translocon. Other current questions concern the interaction of the SecYEG complex with the motor protein SecA. SecA is an ATPase that binds to substrate proteins and delivers them post-translationally to the translocon (for review see Veenendaal et al, 2004). Once bound, SecA undergoes cyclic conformational changes that ultimately result in the ATP-dependant stepwise translocation of the unfolded polypeptide (Schiebel et al, 1991). Despite the fact that SecA has been extensively studied and also crystallized in various conformational states (Hunt et al, 2002; Sharma et al, 2003; Vassylyev et al, 2006; Zimmer et al, 2006), little is known about its association and stoichiometry with the SecYEG complex. SecA exists mostly as a dimer in solution (Woodbury et al, 2002), but its oligomeric state during membrane binding and preprotein translocation is a topic of intense debate; it is proposed to be monomer or dimer depending on the study (Driessen, 1993; Or et al, 2002, 2005; de Keyzer et al, 2005; Jilaveanu et al, 2005). Here again, defining the association between the SecYEG complex and its effector SecA has deep implications for the understanding of the inner working of translocon. Characterizing the interaction and dynamic taking place at a membrane-associated or membrane-embedded protein complex is notoriously difficult. Reconstitution into large lipid bilayer vesicles or analysis in detergent micelles often limits the experimentation or introduces bias in the interpretation of results. Here, we made use of Nanodiscs (Bayburt and Sligar, 2003), an emerging technology that allowed us to prepare nanometer-sized soluble particles containing only one SecYEG complex (Figure 1). Nanodiscs faithfully recreate a lipid bilayer and thus permit investigating the reactivity and interactive nature of SecA with the SecYEG translocon in a near native environment, without use of detergent. Figure 1.Schematic view of a membrane protein complex embedded into the Nanodisc structure (adapted from Bayburt and Sligar, 2003). The image was kindly provided by Ms Kailun Jiang. Download figure Download PowerPoint Results Incorporation of the SecYEG complex into Nanodiscs Nanodisc incorporates membrane proteins into soluble, 10-nm-wide, lipid bilayer structures of controlled size (Bayburt and Sligar, 2003; Denisov et al, 2004). Integral to this technology is a bioengineered 200-amino-acid amphipathic and multihelical membrane scaffold protein (hereafter referred to as MSP), whose hydrophobic faces circumscribe the edges of small lipid bilayer (∼130–160 lipid molecules) and whose polar faces interact with polar aqueous environment (Figure 1). The principle of the self-assembly process and the ratio lipids:MSP forming the Nanodisc structure have been well studied (Denisov et al, 2004), and pioneering studies for proteins containing various seven-transmembrane segments have been reported (Bayburt et al, 2006; Leitz et al, 2006). We used this system to incorporate the SecYEG heterotrimer, which consists of 15 transmembrane helices. Purified SecYEG complex was incubated with an optimized amount of MSPs and total Escherichia coli phospholipids. Following detergent removal, the resulting mixture was fractionated on a calibrated gel filtration column (Figure 2A). The majority of the proteins eluted as two peaks, whereas a minimal amount eluted as larger aggregates in the void volume (V0). SDS–PAGE analysis shows that the larger peak corresponds to an association between the MSP and the SecYEG complex, whereas the second peak corresponds to the MSP alone (Figure 2B). Omission of the MSP from the self-assembly mixture resulted in total aggregation of the SecYEG complex. Thus, the SecYEG complex has been associated with the MSP to form a water-soluble structure characteristic of Nanodiscs containing incorporated target protein (referred to as Nd–SecYEG complex or SecYEG–Nanodisc hereafter). Figure 2.Reconstitution of the SecYEG complex into Nanodiscs. (A) Typical protein elution profile obtained after size-exclusion chromatography of a SecYEG–Nanodisc preparation. (B) The fractions corresponding to elution volume 10–16 ml in (A) were analyzed by SDS–PAGE, followed by Coomassie blue staining of the gel. For molecular size comparison, the SecYEG complex and the MSPs were loaded on thetwo lanes on the left. (C) Sedimentation velocity analysis and c(s) distribution of the Nd–SecYEG complex at 5.6 μM (gray lane) and 9.6 μM (black lane). A small peak is present at lower and higher S values, which represent small amount of free MSP and MSP bound to a dimeric SecYEG complex, respectively. The insets represent the c(M) distribution of the 9.6-μM sample (f/f0=1.4). Download figure Download PowerPoint Stoichiometry and stability of the Nanodisc-reconstituted SecYEG complex To determine the number of SecYEG complex incorporated per Nanodisc, the Nd–SecYEG complex was analyzed by analytical ultracentrifugation (Figure 2C). Sedimentation velocity data obtained at 5.6 and 9.6 μM show a major peak (>90%) that sediments at ∼5.3S. The c(M) distribution (f/f0=1.4; Figure 2C, inset) indicates that the peak corresponds to a molecular weight (MW) of ∼123 kDa, consistent with particles made of two MSPs (∼50 kDa) and one SecYEG complex (∼75 kDa). Such a stoichiometry is also confirmed by blue native PAGE (BN-PAGE) analysis (Figure 3A). The purified SecYEG complex migrates as a population of monomer and dimer (Bessonneau et al, 2002; Figure 3A, lane 1), whereas the Nd–SecYEG complex migrates predominantly as a single band (lane 2), between the SecYEG monomer and dimer. Reconstitution in the absence of the SecYEG complex leads to the formation of empty discs (labeled Nd, lane 3) that migrate slightly above the dimeric MSP, probably owing to the contribution of lipid molecules to the molecular weight of the discs. Depending on the quality of the separation by size-exclusion chromatography, the SecYEG–Nanodisc preparation was sometimes contaminated with free MSP (e.g. Figures 2C and 3A, lane 2). Figure 3.The SecYEG monomer is stably incorporated into the Nanodisc structure. (A) The Nd–SecYEG complex (4 μg, lane 2) was analyzed by BN-PAGE followed by Coomassie blue staining. For comparison, the detergent-soluble SecYEG complex (4 μg, lane 1), the empty discs (2 μg, lane 3) and the MSPs (2 μg, lane 4) were loaded on the same gel. Molecular weight markers: ferritin, 440 kDa; catalase, 232 kDa; aldolase, 158 kDa and BSA 66/132 kDa. (B) The protein samples described in (A) were analyzed by CN-PAGE, followed by Coomassie blue staining of the gel. (C) The [125I]SecYEG and [125I]Nd–SecYEG complexes (each ∼20 000 c.p.m.; ∼20 ng) were incubated at the indicated temperature for 5 min before analysis by BN-PAGE and phosphorimaging. In the presence of 0.1% SDS, both complexes are dissociated. Download figure Download PowerPoint When analyzed by colorless native PAGE (CN-PAGE; Schägger et al, 1994 and Figure 3B), the Nd–SecYEG complex migrates into the gel (lane 2), whereas the detergent-maintained SecYEG complex readily aggregates and remains at the top of the gel (lane 1). The MSP dimers (∼50 kDa, lane 3) and the empty discs (∼50 kDa+lipids, lane 4) are also detected at the expected positions. We note that the sharpness of the bands seems sensitive to the presence of lipids; both the Nd–SecYEG complex and the empty discs, but not the MSPs alone, migrate as smeary bands on native gel. Altogether, the results show that a single SecYEG complex is incorporated in a water-soluble Nanodisc structure. When embedded in the membrane, the SecYEG heterotrimer naturally forms a stable complex (Joly et al, 1994). When extracted from the lipid bilayer with detergents, the SecYEG complex easily dissociates into single subunits (Brundage et al, 1992). This instability can be detected when the purified SecYEG complex is prewarmed before analysis by BN-PAGE (Figure 3C). Temperature leads to the progressive conversion of the SecYEG monomer and dimer into dissociated subunits. The same incubation conditions do not, however, dissociate the Nanodisc-reconstituted SecYEG complex, and the migration pattern of the preparation remains unchanged (Figure 3C). Thus, the disc structure has created a protective environment around the SecYEG complex that mimics that of the natural membrane lipid bilayer. Binding of monomeric SecA onto the monomeric SecYEG complex The SecA protein migrates on CN-PAGE as a single band around 200 kDa, as expected because SecA forms dimers at micromolar concentrations (Woodbury et al, 2002; Figure 4A, lane 1). However, upon incubation with increasing amounts of SecYEG–Nanodisc, the SecA band shifts to a higher position on the gel (lanes 2–5). The same size shift is also observed when the Nd–SecYEG complex is incubated with increasing amounts of SecA (Figure 4B). Identical results were obtained when incubation was performed at room temperature or at 37°C (data not shown). In the control experiment, the SecA protein was incubated with empty discs, but a size shift is not observed either for SecA or for the empty discs (Figure 4A, lanes 6 and 7). Thus, the Nanodisc-reconstituted SecYEG monomer forms a complex with SecA (referred to as Nd–SecYEG–SecA complex hereafter). Formation of the complex is efficient as almost all SecA molecules associate with the SecYEG–Nanodiscs when equimolar amounts are mixed together (Figure 4A, lane 5 or B, lane 6). Figure 4.Binding of the SecA protein onto the SecYEG–Nanodisc. (A) The SecA protein (1 μg, lanes 1–5) was incubated for 5 min at room temperature with the indicated amount of SecYEG–Nanodiscs. As control, the empty discs (2 μg, lanes 6 and 7) were incubated with or without SecA (0.5 μg). The protein samples were separated by CN-PAGE followed by Coomassie blue staining of the gel. (B) The indicated amounts of SecA proteins were incubated with (even lanes) or without (odd lanes) the Nd–SecYEG complex (2 μg) and analyzed by CN-PAGE followed by Coomassie blue staining. Download figure Download PowerPoint The apparent molecular weight of the Nd–SecYEG–SecA complex (i.e. migrating just below the 232 kDa marker, Figure 4) suggests that only one SecA molecule associates per SecYEG–Nanodisc (calculated MW ∼225 kDa). This is a surprising result as SecA is mostly dimeric in aqueous solution and indeed migrates as a dimer on native gels. To ascertain the stoichiometry, the SecA dimeric state was stabilized by oxidation with Cu2+(phenanthroline)3, resulting in the formation of an intra-dimeric SecA disulfide bond (de Keyzer et al, 2005). Incubation with an increasing amount of oxidizing agent converts about half of SecA into covalently linked dimers (Figure 5A, lanes 1–4). On CN-PAGE, both the crosslinked SecA dimers and the native dimers migrate at the same position, around 200 kDa (Figure 5B, lanes 1–4). Next, the preparations of oxidized SecA molecules, made of a mixed population of native and undissociable SecA, were incubated with the SecYEG–Nanodiscs and then separated by CN-PAGE (lanes 5–8). In addition to the Nd–SecYEG–SecA complex, an additional band of higher molecular weight is apparent on the gel. As the intensity of the band increases with the quantity of undissociable SecA dimers (compare Figure 5A, lanes 1–4 with B, lanes 5–8), we conclude that it represents a complex between the SecYEG–Nanodisc and dimeric SecA. Thus, the Nanodisc-reconstituted SecYEG complex binds exclusively monomeric SecA, unless dimeric SecA has been artificially stabilized. Figure 5.Binding of disulfide-linked SecA onto the SecYEG–Nanodisc. (A) The SecA protein was oxidized with Cu2+(phenanthroline)3 at the indicated concentration and aliquots (1 μg) were analyzed by non-reducing SDS–PAGE, followed by Coomassie blue staining of the gel. (B) The same SecA aliquots (labeled SecACP3 on the figure) were analyzed by CN-PAGE, either alone (lanes 1–4) or after incubation with the Nd–SecYEG complex (2 μg; lanes 5–8). (C) Separation of the Nd–SecYEG–SecA complexes by sucrose density centrifugation. The Nd–SecYEG complex was reconstituted in the presence of trace amounts of 125I-labeled SecYEG. The samples contained (1) 70 μg BSA, 70 μg SecA and 30 μg ferritin, (2) [125I]Nd–SecYEG, (3) [125I]Nd–SecYEG+70 μg SecA and (4) [125I]Nd–SecYEG+70 μg of purified cysteine-linked SecA dimer. Samples from fractions 6–20 were analyzed by SDS–PAGE, followed by Coomassie blue staining (panel 1) or phosphorimaging (panels 2–4). The band detected by phosphorimaging is [125I]SecY. Crosslinked SecA dimer and unmodified SecA dimer sediment at the same position on these sucrose gradients (data not shown). Download figure Download PowerPoint This conclusion is also reached by sucrose gradient analysis (Figure 5C). On its own, the [125I]Nd–SecYEG complex sediments in-between BSA and dimeric SecA (fractions 8–10). In the presence of SecA, the [125I]Nd–SecYEG complex sediments in higher-molecular-weight fractions (fractions 11–13), just after the position of dimeric SecA. When the purified cysteine-linked SecA dimer is incubated with the [125I]Nd–SecYEG complex, the peak is further shifted toward higher-molecular-weight positions (fractions 15–17). Dissociation of the SecA dimer onto the monomeric SecYEG complex The above observations strongly suggest that the SecYEG complex causes disassembly of the SecA dimer. To provide addtional evidences, the SecYEG–Nanodiscs and SecA were incubated together with an amine-reactive crosslinking reagent (Figure 6A). In these experiments, either the SecYEG complex reconstituted in Nanodiscs (lanes 1 and 2) or the SecA protein (lanes 3–6) were 125I-radiolabeled. The results show that in both cases, the crosslinking products are identical and migrate at positions in-between the SecA monomer and dimer. Thus, the crosslinker reagent captures the SecYEG complex only with the SecA monomer. Incubation of 125I-labeled SecA with higher amounts of SecYEG–Nanodiscs not only increases the yield of crosslinking products, but also decreases the yield of crosslinked SecA dimers (Figure 6B). This last observation is consistent with the dissociation of the SecA dimer during incubation with the Nd–SecYEG complex. Figure 6.Crosslinking analysis of the Nd–SecYEG–SecA complex. (A) About 2 μg of [125I]Nd–SecYEG (lane 1) was mixed with 1 μg SecA (lane 2) and then incubated with the crosslinker reagent EGS as described in Materials and methods. In the reverse experiment, [125I]SecA (0.5 μg, lanes 3–6) was incubated with unlabeled Nd–SecYEG (2 μg, lane 5) or empty Nanodiscs (2 μg, lane 6) before crosslinking with EGS. Proteins were dissolved in 0.1% SDS and analyzed by BN-PAGE followed by Coomassie blue staining (left panel) and phosphorimaging (right panel). (B) [125I]SecA (0.5 μg) was incubated with the indicated amounts of Nd–SecYEG or empty Nanodiscs. The crosslinking reaction was performed with EGS as described in Materials and methods. Download figure Download PowerPoint The same conclusion is also reached using fluorescence resonance energy transfer (FRET) assays. Two populations of SecA were independently labeled at their carboxy-terminal cysteine residues with either coumarin (DACIA, donor fluorophore) or fluorescein (5-IAF, acceptor fluorophore). The labeled proteins were denatured with urea, mixed together at a 1:2 molar ratio and renatured. The recorded spectra for the heterodimer (Figure 7A, black trace) shows that the fluorescence emission of the donor (dashed black trace) is decreased by 15%, whereas that of the acceptor is increased correspondingly (dashed gray trace). This FRET reflects the dimeric association of the SecA proteins (see also Or et al, 2002). Taking into account the heterodimer formation during refolding (i.e. ∼33% if Poisson distributed at a 1:2 molar ratio), the FRET efficiency was calculated to be 45%. As the SecA monomers and dimers exist in dynamic equilibrium, addition of a five-fold molar excess of unlabeled SecA reduces the energy transfer efficiency (Figure 7B, gray trace). Similarly, addition of a five-fold molar excess of SecYEG–Nanodiscs (Figure 7C, gray trace) strongly abolishes FRET (∼12% increase in donor fluorescence and corresponding decrease in acceptor fluorescence). The intensity of FRET is not changed upon addition of empty discs (Figure 7D). Figure 7.Analysis of the SecA dimer dissociation by steady-state FRET. Excitation was set at 390 nm and emitted light was recorded from 420 to 580 nm. (A) Coumarin/fluorescein-labeled SecA heterodimer (9 μg) (dark trace) compared with coumarin/unlabeled (dashed dark trace) or fluorescein/unlabeled SecA heterodimer (9 μg) (dashed gray trace). The emitted light for coumarin/fluorescein-labeled SecA heterodimer (9 μg, dark trace) was again recorded after the following addition (gray traces). (B) 45 μg of unlabeled SecA; (C) 45 μg of Nd–SecYEG; and (D) 18 μg of empty Nanodiscs. Download figure Download PowerPoint Altogether, the results from native gel electrophoresis, crosslinking analysis, sucrose gradient centrifugation and FRET experiments support the conclusion that the SecYEG complex embedded in Nanodiscs binds the SecA monomer and shifts the SecA monomer–dimer equilibrium toward the dissociated state. The SecY residue R357 is essential for the binding of SecA Previous genetic and biochemical studies indicate that mutations at the conserved residue Arg357, located in the largest cytosolic loop of SecY, prevents SecA ATPase activation (Mori and Ito, 2001; van der Sluis et al, 2006). To test whether the mutation also results in defective SecA–SecYEG association, the SecYEG (R357E) mutant complex was incorporated into Nanodiscs. The results show that the efficiency of complex formation is indeed reduced (Figure 8A, middle panel). Compared with the wild type (right panel), lesser amounts of Nd–SecYEG–SecA complex are formed and, accordingly, higher amounts of SecA remain uncomplexed. Furthermore, when the amino-acyl residues adjacent to R357 are also substituted (sequence R357G358D359 replaced by E357D358P359), the binding of SecA is totally abolished (left panel). Thus, the residue R357 (and the adjacent positions) constitutes a major binding site for SecA. The results also prove that the interaction between the Nd–SecYEG complex and SecA is specific and, as expected, occurs on a SecY loop normally located on the cytosolic face of the membrane. Figure 8.The SecY residue R357 and acidic lipids contribute to the binding of SecA. (A) Wild-type or mutant SecYEG Nanodiscs (1 μg) were incubated for 5 min at room temperature with the indicated amounts of SecA. Samples were analyzed by CN-PAGE followed by Coomassie blue staining of the gel. Incubation at 37°C did not change the results (data not shown). (B, C) The SecYEG complex reconstituted into Nanodisc (1 μg) with the indicated phospholipids was incubated for 5 min at room temperature with the indicated amount of SecA protein and then analyzed by CN-PAGE and Coomassie blue staining. Given the smeary aspect of the Nd–SecYEG–SecA complex on native gel, densitometry analysis was performed on unbound SecA obtained at saturating binding concentration (i.e. 2 μg of SecA in this experiment). The densitometry values were compared with SecA concentration standards run on the same gel. Download figure Download PowerPoint Acidic lipids contribute to the binding of SecA at Nanodisc-reconstituted SecYEG Binding of SecA to inner membrane vesicles depends on phosphatidylglycerol (Lill et al, 1990; Hendrick and Wickner, 1991; Breukink et al, 1992), but the contribution of these lipids to the formation and/or stability of the SecYEG–SecA associations remains unknown. The SecYEG complex was thus reconstituted in Nanodiscs containing different phospholipids; either dioleoylphosphatidylcholine (PC) only or supplemented with 30% dioleoylphosphatidylglycerol (PG). Control experiments included reconstitution without lipids or with total E. coli lipids (made of ∼70% neutral and ∼30% acidic phospholipids). When analyzed by CN-PAGE, these various preparations migrate at different positions depending on the nature of the lipid incorporated (Supplementary Figure S1). The efficiency of reconstitution and the stoichiometry SecYEG:MSP is however not affected by the nature or the absence of lipids (Supplementary Figure S1). In the latter case, it is possible that a few lipid molecules are brought from the purified SecYEG complex itself. Next, the SecYEG–Nanodisc preparations were incubated with SecA and analyzed by CN-PAGE (Figure 8B and C). The results show that the binding capacity of the Nd–SecYEG complex is increased when it contains acidic lipids (Figure 8B, compare left and right panels). When 2 μg of the SecA protein (i.e. saturating concentration) is added to the SecYEG–Nanodiscs containing acidic lipids, the fraction of bound SecA represents ∼55%. This fraction is reduced to ∼18% when SecA is incubated with the SecYEG–Nanodisc containing only neutral lipids. Similarly, the formation of the Nd–SecYEG–SecA complex appears facilitated when the SecYEG–Nanodisc structure includes E. coli lipids (Figure 8C). At saturating concentration (i.e. 2 μg of SecA), the fraction of SecA bound to the Nd–SecYEG complex is ∼62% (right panel), but decreases to ∼20% when the SecYEG–Nanodisc is prepared without additional acidic lipids (left panel). Altogether, the results show that acidic lipids located in the surroundings of the SecYEG complex contribute significantly to the binding of SecA. Activation of SecA at Nanodisc-reconstituted SecYEG The SecA ATPase is stimulated by membrane-embedded or detergent-solubilized SecYEG (Lill et al, 1990; Duong, 2003). Thus, SecA is able to sense the SecYEG complex as an activating ligand. Accordingly, the results show that the endogenous SecA ATPase is stimulated ∼5-fold during incubation with the SecYEG–Nanodiscs (Figure 9A). The stimulation is comparable to the values previously reported when SecA and the purified SecYEG complex are incubated together in detergent solution (Duong, 2003). Neither the MSPs nor the empty discs can activate SecA to the same extent (Figure 9A). Furthermore, the SecYEG–Nanodiscs carrying the SecY mutations analyzed above (i.e. E357 and E357D358P359) do not stimulate the SecA ATPase (Figure 9B), suggesting that the activation requires binding t
Referência(s)