Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Dissecting structures and functions of SecA-only protein-conducting channels: ATPase, pore structure, ion channel activity, protein translocation, and interaction with SecYEG/SecDF•YajC

  • Ying-hsin Hsieh ,

    Contributed equally to this work with: Ying-hsin Hsieh, Ying-ju Huang

    Affiliation Department of Biology, Center for Biotechnology and Drug Design, Georgia State University, Atlanta, GA, United States of America

    ⨯
  • Ying-ju Huang ,

    Contributed equally to this work with: Ying-hsin Hsieh, Ying-ju Huang

    Affiliation Department of Biology, Center for Biotechnology and Drug Design, Georgia State University, Atlanta, GA, United States of America

    ⨯
  • Hao Zhang,

    Affiliation Department of Biology, Center for Biotechnology and Drug Design, Georgia State University, Atlanta, GA, United States of America

    ⨯
  • Qian Liu,

    Affiliation Department of Biology, Center for Biotechnology and Drug Design, Georgia State University, Atlanta, GA, United States of America

    ⨯
  • Yang Lu,

    Affiliation Department of Biology, Center for Biotechnology and Drug Design, Georgia State University, Atlanta, GA, United States of America

    ⨯
  • Hsiuchin Yang †,

    † Deceased.

    Affiliation Department of Biology, Center for Biotechnology and Drug Design, Georgia State University, Atlanta, GA, United States of America

    ⨯
  • John Houghton,

    Affiliation Department of Biology, Center for Biotechnology and Drug Design, Georgia State University, Atlanta, GA, United States of America

    ⨯
  • Chun Jiang,

    Affiliation Department of Biology, Center for Biotechnology and Drug Design, Georgia State University, Atlanta, GA, United States of America

    ⨯
  • Sen-Fang Sui,

    Affiliation State Key Laboratory of Membrane Biology, Center for Structural Biology, School of Life Sciences, Tsinghua University, Beijing China

    ⨯
  • Phang C. Tai

    biopct@gsu.edu

    Affiliation Department of Biology, Center for Biotechnology and Drug Design, Georgia State University, Atlanta, GA, United States of America

    ⨯

Abstract

SecA is an essential protein in the major bacterial Sec-dependent translocation pathways. E. coli SecA has 901 aminoacyl residues which form multi-functional domains that interact with various ligands to impart function. In this study, we constructed and purified tethered C-terminal deletion fragments of SecA to determine the requirements for N-terminal domains interacting with lipids to provide ATPase activity, pore structure, ion channel activity, protein translocation and interactions with SecYEG-SecDF•YajC. We found that the N-terminal fragment SecAN493 (SecA1-493) has low, intrinsic ATPase activity. Larger fragments have greater activity, becoming highest around N619-N632. Lipids greatly stimulated the ATPase activities of the fragments N608-N798, reaching maximal activities around N619. Three helices in amino-acyl residues SecA619-831, which includes the “Helical Scaffold” Domain (SecA619-668) are critical for pore formation, ion channel activity, and for function with SecYEG-SecDF•YajC. In the presence of liposomes, N-terminal domain fragments of SecA form pore-ring structures at fragment-size N640, ion channel activity around N798, and protein translocation capability around N831. SecA domain fragments ranging in size between N643-N669 are critical for functional interactions with SecYEG-SecDF•YajC. In the presence of liposomes, inactive C-terminal fragments complement smaller non-functional N-terminal fragments to form SecA-only pore structures with ion channel activity and protein translocation ability. Thus, SecA domain fragment interactions with liposomes defined critical structures and functional aspects of SecA-only channels. These data provide the mechanistic basis for SecA to form primitive, low-efficiency, SecA-only protein-conducting channels, as well as the minimal parameters for SecA to interact functionally with SecYEG-SecDF•YajC to form high-efficiency channels.

Citation: Hsieh Y-h, Huang Y-j, Zhang H, Liu Q, Lu Y, Yang H, et al. (2017) Dissecting structures and functions of SecA-only protein-conducting channels: ATPase, pore structure, ion channel activity, protein translocation, and interaction with SecYEG/SecDF•YajC. PLoS ONE 12(6): e0178307. https://doi.org/10.1371/journal.pone.0178307

Editor: Eric Cascales, Centre National de la Recherche Scientifique, Aix-Marseille Université, FRANCE

Received: March 3, 2017; Accepted: May 10, 2017; Published: June 2, 2017

Copyright: © 2017 Hsieh et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All data is within the manuscript and supporting information files.

Funding: This work was supported, in part, by NIH grants GM34766 and AI104168, and by NSFC 31230016 and 31370717 (to SFS). This work was also supported by Georgia State University internal funding, including fellowships from Molecular Basis of Diseases.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Most proteins that traverse bacterial cytoplasmic membranes through the Sec-dependent pathway carry a N-terminal hydrophobic signal peptide that is cleaved during, or shortly after, translocation. Biochemical and genetic studies have provided insight into the mechanism of this Sec secretion [1–3]. SecA and its ATPase activity are essential [4–7]. The prevailing model depicts the SecYEG complex as forming the protein-conducting channel with SecA acting as a peripheral ATPase [2, 6], which cycles on and off the cytoplasmic membrane, regulated by SecD-SecF [8, 9]. This model has been reviewed extensively [10–13]. The structure of SecA-SecYEG complex has been reported [14].

There are indications that bacteria can also secrete proteins in the absence of SecYEG both in vitro and in vivo. Blobel’s group first suggested, and subsequent findings have supported such a SecA-dependent pathway that lacks SecYEG [15–21]. Moreover, it has been shown that SecA integrates into membranes [18, 22–26], that it does not necessarily cycle on and off the membrane during protein translocation [23] and that it is an integral part of the protein-conducting channel [15, 22, 23, 26]. SecA can bind to membranes both at SecYEG high-affinity sites and at phospholipid low-affinity sites [23, 25]. We found that, upon binding to anionic phospholipids, SecA forms ring-like pore structures that may be the core of protein-conducting channels [24, 26, 27]. Indeed, SecA alone can elicit ion-channel activity in liposomes and mediate protein translocation, albeit at low efficiency with no signal peptide specificity [16]. A model for SecA-only protein-conducting channel activity has been proposed [15]. SecYEG adds efficiency and specificity [16, 28]. Purified SecYEG-SecDF•YajC transforms SecA-only liposomes channels into high-affinity channels [29]. These two SecA-dependent channels may serve different functions during the cell growth [17].

E. coli SecA was reported to be a cytoplasmic protein with 901 amino acid residues [7] (see S1 Table). X-ray structure of soluble SecA has been reported, albeit with a few uncertain regions, especially in the C-terminal region has been determined [30] (see S1 Fig). It is an ATPase with two separate domains, N68 and C34 [31–33] (see S1 Fig) that contain a high-affinity and a low-affinity nucleotide-binding site (NBD) [8, 32, 34] composed of the Walker A and the Walker B motifs that are present in most ATPases. NBD1 contains the Walker A motif at residues #102–109 and the Walker B motif at #198–210 NBDII contains Walker A at #503–511, and Walker B at #631–653 [32, 34]. X-ray analyses of soluble SecA showed that ATP binds to the interface of NBD1 (defined at #1–220), and NBDII (defined at #377–416), and that both domains are required for ATPase activity [35]. The domain containing the intermolecular regulator of ATP hydrolysis (IRA2) is located within a #421–610 region [[36] see S1C Fig] and does not bind to the nucleotide. It regulates ATP hydrolysis, by controlling ADP release [37]. SecA contains other domains including the preprotein-binding domain (PBXD), the helical-scaffold domain (HSD), the helical-wing domain (HWD) and the C-terminal linker domain (CTL) [1, 38]. The C34 fragment has been proposed to be involved in dimerization [39, 40]; it contains IRA1 domain and the regions involved in SecY or SecB interaction as well as a lipid-binding site and a SecB-binding domain [41].

In this study, we characterized the structural and functional roles of the various SecA domains that participate in ATPase activity, interaction with lipids, formation of pore-ring structures, ion-channels, protein translocation activity, and interactions with SecYEG and SecDF•YajC. We used several N-terminal SecA fragments to define which regions are involved in forming functional SecA-dependent channels. Using deletion-truncated fragments of the 901-residue SecA protein and liposomes in the oocytes recordings, we showed that the Helical Scaffold Domains of aminoacyl residues #620–831 are critical for the formation of functional pore channels in phospholipids, and for interacting with SecYEG-SecDF•YajC to gain higher activity.

Material and methods

Strains and plasmids

MC4100 is derived from Escherichia coli K12 strain (F- lacU169 araD136 relA- rpsL150 frbB5301 deoC7 ptsF25 thi-) which was obtained from J. Beckwith [3]; Strain BA13 (MC4100 supF ts trp am secA13 am zch::Tn10) [4] was obtained from D. Oliver. E.coli strain 773 with ompA deletion is a lab stock [42]. Plasmid encoding SecYEhisG was from F. Duong [43, 44]

Inner membrane vesicles preparation

Wild-type MC4100 membranes, and OmpA-depleted 773 membranes, were prepared as described [42]. SecA-depleted BA-13 membranes were prepared from secAts BA13 mutant. BA13 cells were grown at 30°C to mid-log phase, shifted to 42°C until growth ceased due to SecA depletion. The cells were collected and the BA13 SecA-depleted membranes were prepared by established procedures [42], and subsequently washed with 8 M urea to inactivate residual SecA.

Liposomes preparation

Liposomes of E. coli total lipids (Avanti Lipids) were prepared, as described [16]. The lipids were dried by spin vacuum, re-suspended in 150 mM KCl solution/water and sonicated for 3–5 mins until the solution was clear. Preparations were stored at -80°C and thawed for use only once.

Construction and nomenclatures of N-terminal and C-terminal domain fragments

To construct the C-terminal SecA derivatives, DNA fragments were amplified by PCR (Mastercycler Gradient Pro, Eppendorf) with 5’ primer (TATACATATGCTAATCAAATTGTTAACT) and appropriate 3’ primers using pET5a-SecA. Amplified DNA fragments encoding SecA were cloned into pET5a through NdeI and BamHI. Plasmids carrying SecA derivatives were transformed into SecA deficient mutant cells, BL21.19, at 30°C. Some constructs are shown in Supplementary Information S1 Table. In this paper, except for N68 (N-terminal 68 kD protein, referred as N609 in this paper) and C34 (C-terminal 34 kD protein; SecA610-901), the domains are defined as N-terminal fragments, thus N350 is SecA1-350, N496 is SecA1-496, N609 is SecA1-609, etc. The C-terminal C30 (SecA640-901) and C28 (SecA654-901) were similarly constructed.

Some constructs were not stable; to obtain stable clones the C-terminal residues were mutated, thus domains N629*, N639*, N772* and N798* were single mutations at the C-termini (N629D, F639D, D772R, and D798R). The constructs were verified and confirmed by DNA sequencing in the GSU Biology Core Facility.

Protein purification

E.coli SecA (EcSecA), its derivatives, and pIMBB28 [36] (for His-N68), were over-expressed from plasmids constructed (see S1 Table) and purified from BL21 (λDE3)/pT7-SecA in E. coli BL21 (λDE3) as described [4, 16]. The purity of each fragment was >95% as determined by SDS gels. OmpA and pOmpA were prepared, as previously described [16, 22]. SecDF•YajC was expressed from BL21(λDE3) containing plasmid pET543 (obtained from A. Driessen) and purified as described [29]. SecYEG was prepared from pBAD/secEhisYG (obtained from F. Duong) in C43 strain, as described [16, 45]. Cell-free lysates were passed through a Ni-NTA affinity column (Qiagen), followed by Q-Sepharose cation exchange chromatography (GE Healthcare). SecYEG complex was eluted at 300 to 600 mM NaCl in Tris-HCl, pH 7.9 buffer containing 1% Triton X-100, 10% glycerol and 2 mM DTT, and stored at -80°C in the same buffer until use. Protein amounts were estimated from A280/A260 ratios, and confirmed by Bradford assay as described [16, 29].

In vitro ATPase activity assay

ATPase activity assays were performed with minor modifications, as described previously [46]. For intrinsic ATPase assay, 50 μL reaction mixture contained 1.5 μg of SecA derivatives, 20 μg ovalbumin (optimized for obtaining consistent results), 1.2 mM ATP, 50 mM Tris-HCl (pH 7.6), 20 mM KCl, 20 mM NH4Cl, 2 mM Mg(OAc)2, 1 mM DTT at 40°C for an appropriate time. For Lipid ATPase assays, 3 μg liposomes was incubated at 30°C. ATPase activity was determined by the release of inorganic phosphate detected by the photometric method [46], absorption was measured at 660 nm (SmartSpec Plus, Bio-Rad Laboratories, Inc.). All assays were performed at least in triplicate.

In vitro protein translocation

Translocation of pOmpA into liposomes or membrane vesicles was conducted as described previously [16]. Unless otherwise indicated, the translocation mixtures in 0.1 mL contained 120 μg of liposomes or 4.5 μg of OmpA-depleted, Urea-washed 773 membranes, 1 μg SecA or 1 μg SecA N831 or 0.7 μg SecA N-terminal variants plus 0.3 μg C34, 0.1 μg SecB and 150 ng substrates pOmpA [16]. The mixtures were incubated at 37°C for 30 min. The translocation mixtures were treated with Proteinase K at 400 μg/ml in ice water for 30 min to remove non-translocated pOmpA, and liposomes or membranes were collected by centrifugation. Translocated proteins were detected by immunoblots, as described previously [16, 29].

Xenopus oocyte injection and whole cell recording

Oocytes were collected and prepared from Xenopus laevis (Xenopus Express, Inc) and injected with sample mixtures as described previously [47, 48]; 50 nL sample mixture was injected into oocytes using a Nanoject II injector (Drummond Scientific Co., Broomall, PA). Unless otherwise noted, the amount for each component was 120 ng liposomes, 120 ng SecA, 14 ng pOmpA, 2 mM ATP, and 1 mM Mg2+. The effective concentration of the reagents in 50 nL injected mixtures was estimated based on an average volume of oocytes of 500 nL. The amount for SecYEG and SecDF•YajC used where indicated was 30 ng for each complex [29]. A two-electrode voltage clamp was used to measure the opening of protein-conducting channels in the oocytes as described previously [16, 47, 49]. Current was recorded for 1 min after 3 hours of incubation at 23°C. The inward and outward currents were recorded to measure the net currents.

Atomic force microscopy (AFM)

AFM samples were prepared, as previously described [26], with minor modifications: 1 μg of SecA or its derivatives with or without liposomes was mixed by vortex and incubated on ice for 30 min. AFM images were obtained with a MultiMode VIII Scanning probe microscope (Veeco Instrument Inc., Woodbury, NY) using tapping mode at a scan rate of 0.3 Hz. The data were analyzed by image-processing software (Nanoscope v700) according to the manufacturer’s manual.

Results

Intrinsic and lipid-stimulated ATPase of SecA domains

SecA has two separable, soluble fragments: an N-terminal 68 kDa domain (N68; in this study is referred to as N609, which represents N-terminal SecA1-609, see Supplement Information S1 Table) which possesses high ATPase activity and the C-terminal 34 kD domain (C34) which has been shown to regulate the ATPase activity [36, 50]. The soluble form of intact SecA possesses low intrinsic ATPase activity that is stimulated by anionic phospholipids, and further by SecYEG with precursors [50]. We constructed a series of N-terminal fragments of SecA to determine the location of ATPase activity, lipid-stimulations and other functions. The intrinsic ATPase has been attributed to the nucleotide binding site I (NBDI), which resides within the first 220 aminoacyl residues of SecA (32, 40, see also Supporting Information S1C Fig). SecA N350 which forms partial, pore ring-structures [15] has little ATPase activity that could be stimulated by phospholipids (Fig 1).

thumbnail
Download:
Fig 1. Intrinsic and lipid-stimulated ATPase activities of SecA N-terminal fragments.

The assays are as described in Experimental Procedures.

https://doi.org/10.1371/journal.pone.0178307.g001

We previously found that the first 25-aa N-terminual residues and an EM domain of SecA are necessary for membrane insertion [51, 52], and lipid-ATPase activity [15, 52]. We found that two forms of SecA exist in the membrane [15, 29]. One form (SecAM) that occurs in the presence of phospholipids contains the central M48 tryptic fragment at Glu361.; the other (SecAS) contains a 68-kDa (N609) fragment, which is part of a domain in the soluble SecA [33]. There are two tryptic lipid-specific domains: an N-terminal 36-kDa (N36) and a middle 48-kDa (M48) domain present within the SecAM form [15, 23]. We examined whether various domains of SecA (for domain/fragment names, see nomenclature used in Procedures and S1 Table) have ATPase activity that can be stimulated with phospholipids. Of the SecAS domains (N609 and C34) both of which incorporate into membranes [31], N609 has the higher ATPase activity. Of the SecAM domains, neither N350 nor M48 -starting at Glu361 [15]- has detectable intrinsic ATPase activity, despite each’s having an ATP-binding domain (S1 Fig). Thus the ATPase domain, in addition to NBDI, lies between N350 and N609 (Fig 1). It has been reported that two ATP-binding domains are required to promote translocation [32, 33], and the X-ray structure of SecA indicated that ATP binds to the interface of two NBD sites [35]. We constructed a series of N-terminal SecA domains/fragments associated with C-terminal deletion to determine the potential location of intrinsic ATPase and lipid-stimulated ATPase activities. While most purified fragments interacted with lipids, the critical N-terminal domain for ATPase was around N492, which has 30% higher intrinsic activity than that of the intact SecA. Even so, the lipid-stimulated ATPase activity was significantly less, when compared to intact SecA (1.4-fold compared to 4-fold with intact SecA under the conditions tested). The intrinsic ATPase activity of the N-terminal fragments increased up to N632 (Fig 1). The soluble N609 domain had a high intrinsic ATPase activity that increased when interacting with lipids (the reported lack of stimulation by lipids is due to the presence of His-N609, data not shown). Maximal lipid-stimulated ATPase activity was found to be around domain-fragment N619; successive fragments remained high up to N798 (Fig 1). ATPase activities decreased from N798—N831, with the intact SecA having the lowest activity of all (Fig 1), presumably due to regulatory inhibition by interactions with the C-terminal domain [36, 53].

Formation of ring-pore structure with liposomes observed in AFM

SecA forms pore-ring structures [26] when interacting with anionic phospholipids. The two lipid-specific domains N350 and M48 form partial ring structures in the presence of phospholipids that are smaller than the complete SecA ring-structure [15]. As the length of SecA fragments was extended beyond N350, the partial pore-structure was lost at around N462; soluble fragments N609—N628 individually interact with lipids and have high ATPase activity, but form no pore structures (Fig 2A). Pore-ring structures could be observed with fragment N639 and larger (Fig 2A), but higher efficiency with N640-N668 with liposomes. These pore structures were different from the partial pore structures observed with N350 or M48 [15]. However, such structures were smaller than those with the full SecA (Fig 2B).

thumbnail
Download:
Fig 2. AFM structure image of SecAs.

(A) AFM images formed by SecA/SecA fragments. (B) Normal distribution of radius of ring-like structures formed by SecA/SecA N-fragments. Plots were generated by R program, adopted from http://www.r-projet.org (a free software for statistical computing and graphics), based on 100 random samples with similar means and variances of each measured datum. The efficiency of forming pore-ring structure for N640 was about 40%, for N643 or larger >70%, and for N668 >90%.

https://doi.org/10.1371/journal.pone.0178307.g002

SecA-only ion-channel and protein translocation activity

We previously identified a SecA-only protein-conducting channel activity in oocytes and in vitro protein translocation, and determined its characteristics and the differences from SecA-dependent channels containing SecYEG [16]. We proposed that SecA contains two binding sites for membranes and protein translocation: one with low-affinity phospholipids SecA-only and the other with high-affinity SecYEG [15, 16, 28]. SecA-only channels are less efficient and require additional ATP to elicit even low levels of channel activity [16]. In this study, we determined the SecA-only channel activity of various N-terminal fragments of SecA when associated with liposomes in the oocytes for ion channel activity. The the N350- or M48-liposomes, which form only partial ring-structures [15], had no channel activity (data not shown). Even though fragments N639-N668 formed relatively small ring-pore structures with liposomes, they had no ion channel activity. N738-liposomes, however, began to show a little channel ion current activity with N772-liposomes demostrating about 50% channel activity (with a lower expression rate), while N798-liposomes possessed substantial channel activity (Fig 3A). N828 and larger fragments showed almost the full channel activity of intact SecA-liposome channels.

thumbnail
Download:
Fig 3. Function of SecA-only liposomes.

(A) Channel activity. (B) Protein translocation. Assays as described in Experimental procedures.

https://doi.org/10.1371/journal.pone.0178307.g003

In addition to eliciting ion channel activity, SecA-liposomes promote protein translocation [16]. We therefore determined whether there is any correlation between SecA-only ion-channel activity and protein translocation activity. In contrast to the channel activities elicited by fragments larger than N772, there was little to no translocation of pOmpA precursors into liposomes up to domain N828; with N831 being the lowest SecA fragment tested that promoted detectable translocation in liposomes (Fig 3B). The 3 residues SecA828-831 form part of a hydrophobic helix (S1D Fig), and appeared to be critical for SecA-only channels to translocate proteins.

SecA domains interacting with SecYEG-SecDF•YajC for channel activity and protein translocation

SecA-SecYEG-SecDF•YajC channels are specific and efficient [16]. We further defined the critical SecA domains’ interactions with SecYEG-SecDF•YajC for ion channel and translocation activity (Fig 4). There was no channel activity in N350, N609, and N619 in the presence of liposomes containing SecYEG or SecYEG-SecDF•YajC, the latter two having been previously shown to possess no channel activity by themselves [16]. The liposomes of N629*, N639* and N640 fragments when complexed with SecYEG had some channel activity, though with less expression efficiency (Fig 4A). There were substantial channel activities for N643-N649 fragments with SecYEG and SecDF•YajC, indicating that the critical SecA residues that interact with SecYEG and SecDF•YajC for channel activity may be located in the SecA643-649 aminoacyl residues. The N668 domain, which by itself has no channel activity with liposomes, has full channel activity with SecYEG and SecYEG-SecDF•YajC (Fig 4A and 4B). Channel activity was similarly observed with domains N643, N649, or N668 together with SecA-depleted membranes (Table 1, and data not shown). Although ion channel activity was restored by the reconstitution of the SecA domain fragments with SecYEG-SecDF•YajC, ionic current and expression efficiency in the oocytes were slightly less effective when compared to the SecA-depleted membranes containing SecYEG-SecDF•YajC.

thumbnail
Download:
Fig 4. Function of SecA-liposomes with SecYEG/SecDF-YajC.

(A) SecA-domain-SecYEG channel activity. (B) SecA-domain with SecYEG-SecDF-YajC Channel activity. (C) Translocation activity in membranes: 1 μg SecA/fragment and 4.5 μg urea treated OmpA-depleted 773 membranes or 120 μg liposomes were used. The translocation in 100 μL were conducted under 37C for 30 min and analyzed as in Experimental procedures.

https://doi.org/10.1371/journal.pone.0178307.g004

thumbnail
Download:
Table 1. Summary of critical SecA domains for structure and function with lipids.

https://doi.org/10.1371/journal.pone.0178307.t001

In order to determine the ion current activity of various domain-fragments with SecYEG and SecYEG-SecDF•YajC we extended the C-terminual amino acid residues to fragments N738*, 772, 798*, 828, and 831. Having shown that the fragment N668 had full ion channel activity with SecYEG-liposomes, increasing the length of SecA fragments to N738*, N772 and then N798* resulted in a significant decrease in channel activity before it was regained in fragments N828 and N831 (Fig 4A and 4B), N831 was able to fully express channel activity, indicating the special function of residues 828–831 which form part of the third helix (S1D Fig). Thus, while amino acids #668–798 appeared to be important for liposomes interacting with SecYEG; their activity is enhanced by residues 828–831.

In view of the conformational changes of SecA that are induced by phospholipids, we tested and correlated various SecA domains interacting with liposomes for their pOmpA translocation activity with SecYEG and SecDF-YajC. As noted earlier, there is no translocation activity with liposomes alone with SecA domains smaller than N828 (Fig 3B). Addition of SecYEG-SecDF•YajC did allow some translocation activity for SecAN828 (~10%), but addition of SecYEG-SecDF•YajC with SecAN831 increased translocation greatly, almost to the same extents as with the full length SecA and intact membranes (Fig 4C). These data again indicate the importance of the SecA828-831 region for interacting with SecYEG-SecDF•YajC and promoting protein translocation.

Complementation of inactive N- and C-SecA framents for structure and functions

Our data indicate that the SecA aminoacyl residues #609–831 are critical for pore-formation, channel activity, protein translocation and interaction with SecYEG. The X-ray structures [40] show that this domain contains 3 helices (S1D Fig): the long Helical Scaffold Domain (#619–668), which consists of two subdomains, HSDI (#619–639) and HSDII (#640–668), the latter of which forms the pore-structure, and possesses channel activity. HSDI has no such capability, but possibly provides the ability to reconstitute active channels through interactions with otherwise inactive C-terminal domains. The C34 domain (SecA610-901), which contains the 3 helices and the extreme C-terminal 70 aminoacyl residues, has been characterized for its ability to down-regulate ATPase activity [IRAI; [8, 36]]. We examined the ability of this domain to complement the various non-functional N-terminal domains for pore formation, channel activity and protein translocation.

SecA-liposome domain reconstitution.

Fragments N609 and N619, which by themselves do not, form pore structures in liposomes (Fig 2), do so in combination with C34 (SecA610-668), though with low efficiency; they are similar in size to those formed by intact SecA (Fig 5A and 5B), N609 with C34 has no ion channel activity, which serves as internal negative control for complementation with C34 (Fig 6A). N619 and C34 overlaps residues #610–619 and together have only a little ion-channel activity; reconstitution of C34 with fragments N629-640 yields more. Reconstitution with N643-N668 yields as much channel activity as those with SecA-only-liposomes (Fig 6A). Protein translocation activities of C34 also increased with longer N-terminal regions in SecA-only reconstituted liposomes, reaching more than 60% of the activity of intact SecA (Fig 6B). (Such translocation activity is comparable to that with SecYEG-SecDFC which is about 50% of intact SecA activity, Fig 6B).

thumbnail
Download:
Fig 5. Complementation of SecA N-domains for AFM ring-structures.

Same amounts of proteins were used: molar ratios of C34: N-fragments were about 2:1. Full length SecA was used for comparison. (A) AFM images with mean sizes and SD. (B) Size distribution as analyzed in Fig 2B. No complementation of pore structures with ovalbumin instead of C34 (data not shown).

https://doi.org/10.1371/journal.pone.0178307.g005

thumbnail
Download:
Fig 6. Functional complementation of SecA fragments with C-terminal fragment C34.

(A) Channel activity in SecA domains reconstituted with C34, SecA610-901. (B) Translocation activity in SecA domains reconstituted with C34. N609 together with C34 have no activity, and could serve as internal negative controls. The reactions were conducted with liposomes with SecYEG or SecYEG-SecDFC as indicated. 100% translocation activity is SecA with SecYEG-DFC.

https://doi.org/10.1371/journal.pone.0178307.g006

The C34 domain overlaps various N619-N668 fragments, e.g. it overlaps almost 60 aminoacyl residues of N668. We examined the interaction of N668 with smaller C-terminal constructs, including C30 (SecA640-901) and C28 (SecA654-901). Interactions with both C-terminal constructs yielded active SecA-only channel activity; the interaction with C28 (with only a 13 aminoacyl residue overlap) yielded almost complete SecA-alone channel activity (Fig 7A). These findings show that the region SecA610-653 of C34 is not necessary for complementation, and that small overlapping regions yield channels activities equivalent to that of SecA-alone.

thumbnail
Download:
Fig 7. Complementation with C-terminal domains for activities.

(A) Liposomes channel activity. N643 or N668 alone has no channel activity (B) Membrane protein translocation. C34 fragment: SecA610-901. C28 fragment: SecA662-901. C30 fragment: SecA 640–901. None of SecA fragments alone has any translocation activity.

https://doi.org/10.1371/journal.pone.0178307.g007

SecA fragments reconstitution with SecYEG-SecDF•YajC.

We examined interactions of SecA fragments and liposomes with SecYEG-SecDF•YajC, and found some synergistic increases in channel activities with fragments up to N640. Domains in HSDII #643–668, which had high channel activity with C34, did not increase channel activity (data not shown), indicating that large overlaps in the HSD helix do not enhance interactions with SecYEG-SecDF•YajC or with intact membranes. In contrast, translocation activities of reconstituted C34 and N-terminal SecA domains were much more pronounced (Fig 6B). There were substantial translocation activities for fragments N619- and N629- liposomes (but not N609) with SecYEG-SecDF•YajC and C34, while N640-N668 reached maximal levels when reconstituted with SecYEG-SecDF•YajC and C34, attaining translocation activity of about 50% activity found for intact SecA (Fig 6B).

We also examined the translocation effects of C-terminal overlapping residues in interaction with native SecA-depleted membranes containing SecYEG-SecDF•YajC. There were translocation activities observed in all C34, C30 and C28 fragments in combination with N643 and N668, with the longer C34 having a slightly higher translocation activity (Fig 7B). There was no translocation activity when combined with no overlapping region, i.e. N609. The overlapping regions, therefore, did appear to have some effects on translocation.

Discussion

E.coli SecA, which contains 901 amino acids, interacts with multiple components of the Sec protein translocation system. In this study, we defined which of the SecA domains are important for interaction with lipids, for ATPase activity, for ring-pore formation, for ion channel activity, and for protein translocation. We also defined which regions interact with SecYEG-SecDF-YajC (Table 1). In liposomes alone, SecA1-831 is minimally required for protein translocation;, SecA1-772 for ion channel activity, SecA1-639 for pore formation, and SecA 1–609 for ATPase activity. SecA1-649 and SecA1-828 interact with SecYEG-SecDF•YajC yielding higher efficient channel and translocation activities, respectively (Table 1). Thus, we have established which SecA domains are required for each of the two SecA-dependent protein translocation pathways: the low-affinity, low-efficiency, SecA-only protein-conducting channels, and the high-affinity, high-efficiency SecA-SecYEG-SecDF•YajC protein-conducting channels [15, 16, 29]. This work provides evidence for which domains are critical for structures and function, particularly the three helical domain region of SecA (SecA619-831), which appears to coordinate key features of these processes (S1 Fig and Table 1).

The soluble N609 domain has high intrinsic ATPase activity and can interact with lipids to yield higher activity. The soluble N609 domain containing NBDI and probably a partial NBDII [32, 34] also has high intrinsic ATPase activity which increases upon interaction with lipids. N-terminal deletions dramatically affect this activity [39, 52]; and a point mutation, F586L, lost the lipid-stimulated ATPase activity of N619 (Data not shown). The long helix (S1C and S1D Fig), which contains a putative NBD II Walker B site (#631–653, Ref 32; S1C Fig), drastically changes the interaction with lipids. The region of SecA from residues #621–831 is particularly important for forming structures and functions with lipids. It consist of three helices (S1C and S1D Fig): 1) the helical Scaffold Domain HSD (N619-668), 2) HSDI, N621-640 which forms fully functional channel activity with membranes, HSDII, N641-668 with SecYEG-SecDF•YajC, HWD (N762-782) and RAI (N810-831). Table 1 highlights which domains that are critical for forming pore structures with liposomes (N629), for channel activity with membranes (N640), and for interactions with SecYEG (N649), SecDF•YajC (N643), liposome-only channels (N798) and translocation (N829). These regions appear to be critical for forming pore structures in the liposomes. N640 and larger domains up to N668 form small ring-pore structures (Fig 2), which by themselves have no ion channel activity in the liposomes (Table 1). Domains between N649-N668 yield higher ion channel activity when reconstituted with SecYEG-SecDF•YajC (Table 1). In contrast, differences in protein translocation activities are relatively small (#828–831) for SecA-only and SecA-SecYEG-SecDF•YajC channels.

SecA can exist as a soluble hydrophilic protein, yet it has many regions that interact with lipids, allowing it to integrate into liposomes and membranes [15, 22, 23, 26, 31, 54, 55]. Previous studies revealed several sites for liposomal binding: one is located within the region encompassed by the amino acids #1–22 [52], #101–118 [51] and two more [25] that form partial pore structures [15, 22]. It has been proposed that SecA contains two lipids binding-sites located at both the N-terminal (1–100) and C-terminal (600–900) ends of the protein [31, 56, 57]. The C-terminal has been reported to contain lipid binding sites [41, 55, 58, 59]. Keller [60], predicted several SecA-lipid binding sites (SecA1-21, SecA14-33, SecA43-60, SecA66-90, SecA108-125, SecA370-395, SecA400-488 SecA593-614 SecA635-660 SecA804-822 SecA865-882, SecA877-895). Thus, many SecA domains interact with lipids and, in so doing, affect ATPase activity, structure and function. These interactions are dynamic. Our results are compatible with these previous findings, and define more precisely which regions determine ring-pore-structure formation, that form ion channels, as well as which regions interact with SecYEG-SecDF•YajC mediating protein translocation.

There are two membrane-binding sites in SecA in the membranes: one high-affinity, with SecYEG, and the other low-affinity with liposomes alone [2, 16, 32, 33, 61, 62]. Translocation at each site is SecA-dependent and their efficiency and specificities vary with SecA’s interaction with SecYEG-SecDF•YajC, which SecA dependent [16, 28, 29]. Using SecA-deletion fragments, we are able to map the critical domains responsible for the channel activity of SecA-only and/or SecA-SecYEG-SecDF•YajC liposomes. SecA amino acid residues #640–649 are critical for SecYEG interactions (Table 1; S1D Fig). These observations are consistent with earlier site-directed, in vivo photo-crosslinking studies [63–65] that identified several amino acids including #640 and #661 of SecA as being essential for SecYEG interactions. The crystal structures of SecA-SecYEG complex also indicates the importance of the SecA long HSD interacting with SecY [14], and #640–661 are structurally critical for SecY interactions. Our data further support the importance of this region for the SecA-SecYEG functional interactions. However, the SecA-SecYEG-SecDF•YajC liposomes failed to generate ionic currents as high as those observed with E. coli membranes containing SecYEG-SecDF•YajC, an inability that was also observed earlier with pPhoA [28]. This may suggest that the translocation machinery components in the membrane have more than just SecA and SecYEG-SecDF•YajC. The crystal structure of SecDF has been reported [66]. SecDF is involved in the stabilization of SecYEG, and maintenance of a proton motive force to enhance protein translocation [44, 66–69]. Other accessory proteins also play important roles in translocation channel activity. Randall’s group has identified several SecA domains that interact with the molecular chaperone SecB [62, 63].

We previously proposed that there are two SecA-dependent protein-conducting channels: the SecA-SecYEG-SecDF•YajC high-affinity protein conducting channels, and low-affinity, less specific SecA-only channels [15, 16, 28, 29]. This work supports that proposal and defines which SecA domains interact with lipids to form SecA-only channels. These interactions provide a basis for there being different structures and functions for SecA-only channels. The domains required for pore formation are different from those required for the ion-channel and protein-translocation activities. Identification of different domains for ATPase, pore structure, channel activity and protein translocation support the idea that SecA functions at least as a dimer [15, 27, 39, 56, 70, 71] and refs. therein) in the SecA-only channels. Interactions with SecYEG-SecDF•YajC require fewer SecA residues. This work further refines the SecA domains that interact with SecYEG-SecDF•YajC to form high-efficient and signal-specific channels. Such (SecA)-SecYEG channels are probably more important during the exponential phase of growth [17, 28]. There is a notable surfeit of SecA in stationary phase [17]. Although it is not clear how our findings relate to in vivo, we speculate that SecA-only channels play a physiological role in stationary phase [16, 17, 29], as well as secretion of cytoplasmic proteins that lack a signal peptide [72]. SecA is required for biogenesis and co-translational insertion of certain membrane proteins [73–76]. SecA binds to ribosomes [77, 78] and competes (in a mutually exclusive manner) with SecYEG for such attachment [79]. Ribosomes might associate with more with SecA-only channels than with SecYEG channels for co-translational secretion, especially for proteins that lack signal peptides. Many cytoplasmic membrane proteins lack cleavable signal sequences, and so may not be recognized by SecYEG channels, the components of which also lack a signal peptide, and presumably be placed in the membrane through SecA-only channels. Thus, SecA-only channels may involve co- and post-translational translocation, Their competitive binding to ribosomes may explain why there is more SecA than SecYEG in bacterial cells.

Supporting information

S1 Table. E. coli SecA protein sequence and the construct domain sequences.

Protein sequence of EcSecA derived from nucleotide sequence [33]. The residues for three critical helices (see S1C and S1D Fig) are color high-lighted. The various domains were constructed, over-expressed and purified. The extreme C-termini of N629 and N639 were mutated with D for stable expression of the constructed domains. The domains are presented as residues from SecA, thus N609 refers to SecA1-609, or C-terminal domains, C34 refers to SecA610-901.

https://doi.org/10.1371/journal.pone.0178307.s001

(DOCX)

S1 Fig. Structure and hydrophobicity of E.coli SecA.

A. Hydrophilicity plot of SecA sequence [34] (from proScale, ExPASy Bioinformatics Resource Portal).

B. Multiple lipid-binding domains [15, 22, 25, 31, 33, 51, 52]. N68 (N609), C34 (# 610–901), N350 and M48 (SecA361-798) are trypsin fragments; the latter 2 are lipid-specific [15, 22].The lower bars: known and predicted SecA lipid-binding domains [60]: (SecA1-21, SecA14-33, SecA43-60, SecA66-90, SecA108-125, SecA370-395, SecA400-488 SecA593-614 SecA635-660 SecA804-822 SecA865-882, SecA877-895).

C. Linear presentations of critical SecA domains and Subdomains: Nucleotide binding domain (NBDI and II NBDI: A1:102–109; B1:198–210; NBDII: A2:503-511B2: 631–653 [34]; Preprotein binding domain); α-helical wing domain (HWD; a long helical scaffold domain (HSD) #621–668 containing two subdomains:# 621–640 (Orange) important for forming pores and 641–668 (Green) critical for SecYEG interaction [14, 32, 64].1 deep-blue helix RA1 (# 756–788) for channel activity; 1 red-helix (#802–829) for protein translocation.

D. Simulated EcSecA X-ray ribbon structure [30]. (Protein Data Bank #2FSF; only #12–831 available; Pink, N-terminal helix #1–11; precursor binding domain, (PPXD) and C-terminal domains CTD are not available). E. coli SecA monomer 3D ribbon structure predicted by Pymol software (The PyMOL Molecular Graphics System, Version 1.7.4 Schrödinger, LLC.).

Left panel: SecA ribbon structure of 3 highlighted critical helix domains. The 3 helical domains #609–831 form hydrophobic interactions and are probably critical for SecA functions. Right panel: Zoom in simulated 3D structure of three critical helices with interacting hydrophobic residues. Green: 651-667aa, Blue: 762-782aa, and Red: 810-829aa. The hydrophobic amino acids are marked on three helices. The I695 is close to F811, M814, L815 and L818 after 180 degree turn on 3D structure.

https://doi.org/10.1371/journal.pone.0178307.s002

(DOCX)

Acknowledgments

We thank all colleagues who provided the strains and plasmids over the years in our work. We thank Irene Weber for discussion and comments, Tingting Shen and Chanda Alexander for technical support. This work was supported, in part, by NIH grants GM34766 and AI104168 (to PCT), and by NSFC 31230016 and 31370717 (to SFS). YJH and HSH were fellows of Molecular Basis of Diseases Program at Georgia State University. We thank Ping Jiang of the Biology for DNA sequencing in DNA/Protein Core facility, which is supported, in part, by the Georgia Research Alliance.

This article is dedicated to John Ingraham for his insights and unfailing support for the considered ‘farfetched’ view that SecA-only channels may function physiologically in the membranes.

Author Contributions

  1. Conceptualization: PCT CJ SFS.
  2. Formal analysis: YHH YJH HZ QL YL HCY.
  3. Funding acquisition: PCT CJ SFS.
  4. Investigation: YHH YJH HZ QL YL HCY.
  5. Methodology: YHH YJH HZ QL YL HCY.
  6. Resources: PCT CJ SFS JH.
  7. Software: SFS JH.
  8. Writing – original draft: YHH PCT HCY JH.
  9. Writing – review & editing: YHH PCT JH.

References

  1. 1. Driessen AJ, Nouwen N. Protein translocation across the bacterial cytoplasmic membrane. Annu Rev Biochem. 2008;77:643–67. Epub 2007/12/15. pmid:18078384
  2. 2. Manting EH, Driessen AJ. Escherichia coli translocase: the unravelling of a molecular machine. Molecular microbiology. 2000;37(2):226–38. pmid:10931320
  3. 3. Silhavy TJ, Beckwith J. Isolation and characterization of mutants of Escherichia coli K12 affected in protein localization. Methods Enzymol. 1983;97:11–40. Epub 1983/01/01. pmid:6228708
  4. 4. Cabelli RJ, Chen LL, Tai PC, Oliver DB. Seca Protein Is Required for Secretory Protein Translocation into Escherichia-Coli Membrane-Vesicles. Cell. 1988;55(4):683–92. pmid:2846186
  5. 5. Chen L, Tai PC. ATP is essential for protein translocation into Escherichia coli membrane vesicles. Proceedings of the National Academy of Sciences of the United States of America. 1985;82(13):4384–8. pmid:2861605
  6. 6. Chen L, Tai PC. Evidence for ATP involvement in cotranslocational protein translocation. Nature. 1987;328:164–6. pmid:3299104
  7. 7. Oliver DB, Beckwith J. Identification of a new gene (secA) and gene product involved in the secretion of envelope proteins in Escherichia coli. J Bacteriol. 1982;150(2):686–91. Epub 1982/05/01. PubMed Central PMCID: PMC216417. pmid:6279567
  8. 8. Economou A, Pogliano JA, Beckwith J, Oliver DB, Wickner W. SecA membrane cycling at SecYEG is driven by distinct ATP binding and hydrolysis events and is regulated by SecD and SecF. Cell. 1995;83(7):1171–81. Epub 1995/12/29. pmid:8548804
  9. 9. Economou A, Wickner W. SecA promotes preprotein translocation by undergoing ATP-driven cycles of membrane insertion and deinsertion. Cell. 1994;78(5):835–43. pmid:8087850
  10. 10. Allen WJ, Corey RA, Oatley P, Sessions RB, Baldwin SA, Radford SE, et al. Two-way communication between SecY and SecA suggests a Brownian ratchet mechanism for protein translocation. eLife. 2016;5:e15598. pmid:27183269
  11. 11. Bauer Benedikt W, Shemesh T, Chen Y, Rapoport Tom A. A “Push and Slide” Mechanism Allows Sequence-Insensitive Translocation of Secretory Proteins by the SecA ATPase. Cell. 2014;157(6):1416–29. pmid:24906156
  12. 12. Lycklama a Nijeholt JA, Driessen AJM. The bacterial Sec-translocase: structure and mechanism. Philosophical Transactions of the Royal Society B: Biological Sciences. 2012;367(1592):1016–28.
  13. 13. Park E, Rapoport TA. Mechanisms of Sec61/SecY-Mediated Protein Translocation Across Membranes. Annual Review of Biophysics. 2012;41(1):21–40.
  14. 14. Zimmer J, Nam Y, Rapoport TA. Structure of a complex of the ATPase SecA and the protein-translocation channel. Nature. 2008;455(7215):936–43. pmid:18923516
  15. 15. You Z, Liao M, Zhang H, Yang H, Pan X, Houghton JE, et al. Phospholipids induce conformational changes of SecA to form membrane-specific domains: AFM structures and implication on protein-conducting channels. PLoS One. 2013;8(8):e72560. Epub 2013/08/27. PubMed Central PMCID: PMC3745498. pmid:23977317
  16. 16. Hsieh Y-h, Zhang H, Lin B-r, Cui N, Na B, Yang H, et al. SecA alone can promote protein translocation and ion channel activity. J of Biol Chem. 2011;286(52):44702–9.
  17. 17. Yang CK, Lu CD, Tai PC. Differential expression of secretion machinery during bacterial growth: SecY and SecF decrease while SecA increases during transition from exponential phase to stationary phase. Current microbiology. 2013;67(6):682–7. pmid:23852076
  18. 18. Watanabe M, Nicchitta CV, Blobel G. Reconstitution of protein translocation from detergent-solubilized Escherichia coli inverted vesicles: PrlA protein-deficient vesicles efficiently translocate precursor proteins. Proc Natl Acad Sci U S A. 1990;87(5):1960–4. Epub 1990/03/01. PubMed Central PMCID: PMC53604. pmid:2408048
  19. 19. Yang YB, Lian J, Tai PC. Differential translocation of protein precursors across SecY-deficient membranes of Escherichia coli: SecY is not obligatorily required for translocation of certain secretory proteins in vitro. J Bacteriol. 1997;179(23):7386–93. Epub 1997/12/11. PubMed Central PMCID: PMC179689. pmid:9393703
  20. 20. Yang YB, Yu N, Tai PC. SecE-depleted membranes of Escherichia coli are active. SecE is not obligatorily required for the in vitro translocation of certain protein precursors. J Biol Chem. 1997;272(21):13660–5. Epub 1997/05/23. pmid:9153216
  21. 21. Baars L, Wagner S, Wickstrom D, Klepsch M, Ytterberg AJ, van Wijk KJ, et al. Effects of SecE depletion on the inner and outer membrane proteomes of Escherichia coli. J Bacteriol. 2008;190(10):3505–25. Epub 2008/02/26. PubMed Central PMCID: PMC2395001. pmid:18296516
  22. 22. Chen X, Brown T, Tai PC. Identification and characterization of protease-resistant SecA fragments: secA has two membrane-integral forms. J Bacteriol. 1998;180(3):527–37. Epub 1998/02/11. PubMed Central PMCID: PMC106918. pmid:9457854
  23. 23. Chen X, Xu H, Tai PC. A significant fraction of functional SecA is permanently embedded in the membrane. SecA cycling on and off the membrane is not essential during protein translocation. J Biol Chem. 1996;271(47):29698–706. Epub 1996/11/22. pmid:8939903
  24. 24. Chen Y, Tai PC, Sui SF. The active ring-like structure of SecA revealed by electron crystallography: conformational change upon interaction with SecB. J Struct Biol. 2007;159(1):149–53. Epub 2007/04/10. PubMed Central PMCID: PMC2691388. pmid:17419072
  25. 25. Kim YJ, Rajapandi T, Oliver D. SecA protein is exposed to the periplasmic surface of the E. coli inner membrane in its active state. Cell. 1994;78(5):845–53. Epub 1994/09/09. pmid:8087851
  26. 26. Wang HW, Chen Y, Yang H, Chen X, Duan MX, Tai PC, et al. Ring-like pore structures of SecA: implication for bacterial protein-conducting channels. Proc Natl Acad Sci U S A. 2003;100(7):4221–6. Epub 2003/03/19. PubMed Central PMCID: PMC153074. pmid:12642659
  27. 27. Wang H, Na B, Yang H, Tai PC. Additional in vitro and in vivo evidence for SecA functioning as dimers in the membrane: dissociation into monomers is not essential for protein translocation in Escherichia coli. J Bacteriol. 2008;190(4):1413–8. Epub 2007/12/11. PubMed Central PMCID: PMC2238208. pmid:18065528
  28. 28. Zhang H, Hsieh YH, Lin BR, Yu L, Yang H, Jiang C, et al. Specificity of SecYEG for PhoA precursors and SecA homologs on SecA protein-conducting channels. Biochem Biophys Res Commun. 2013;437(2):212–6. Epub 2013/06/25. PubMed Central PMCID: PMC3740335. pmid:23791875
  29. 29. Hsieh Y H, Zhang H. Yang C. Jiang S-F. Sui, and Phang C. Tai. Reconstitution of SecA-dependent protein-conducting channels as functionally efficient as in membranes: Transformation of low-affinity SecA-liposome channels to high-affinity SecA-SecYEG-SecDF.YajC. Biochem Biophys Res Commun. 2013;431:388–92. pmid:23337498
  30. 30. Papanikou E, Karamanou S, Economou A. Bacterial protein secretion through the translocase nanomachine. Nature reviews Microbiology. 2007;5(11):839–51. pmid:17938627
  31. 31. Dapic V, Oliver D. Distinct membrane binding properties of N- and C-terminal domains of Escherichia coli SecA ATPase. J Biol Chem. 2000;275(32):25000–7. Epub 2000/06/03. pmid:10835419
  32. 32. Mitchell C, Oliver D. Two distinct ATP-binding domains are needed to promote protein export by Escherichia coli SecA ATPase. Molecular microbiology. 1993;10(3):483–97. pmid:7968527
  33. 33. Price A, Economou A, Duong F, Wickner W. Separable ATPase and membrane insertion domains of the SecA subunit of preprotein translocase. J Biol Chem. 1996;271(49):31580–4. Epub 1996/12/06. pmid:8940175
  34. 34. Schmidt MG, Rollo EE, Grodberg J, Oliver DB. Nucleotide sequence of the secA gene and secA(Ts) mutations preventing protein export in Escherichia coli. J Bacteriol. 1988;170(8):3404–14. PubMed Central PMCID: PMC211308. pmid:2841285
  35. 35. Hunt JF, Weinkauf S, Henry L, Fak JJ, McNicholas P, Oliver DB, et al. Nucleotide control of interdomain interactions in the conformational reaction cycle of SecA. Science. 2002;297(5589):2018–26. pmid:12242434
  36. 36. Karamanou S, Vrontou E, Sianidis G, Baud C, Roos T, Kuhn A, et al. A molecular switch in SecA protein couples ATP hydrolysis to protein translocation. Molecular microbiology. 1999;34(5):1133–45. pmid:10594836
  37. 37. Papanikou E, Karamanou S, Baud C, Sianidis G, Frank M, Economou A. Helicase Motif III in SecA is essential for coupling preprotein binding to translocation ATPase. EMBO Rep. 2004;5(8):807–11. Epub 2004/07/24. PubMed Central PMCID: PMC1299117. pmid:15272299
  38. 38. Driessen AJ, Manting EH, van der Does C. The structural basis of protein targeting and translocation in bacteria. Nat Struct Biol. 2001;8(6):492–8. Epub 2001/05/25. pmid:11373615
  39. 39. Das S, Stivison E, Folta-Stogniew E, Oliver D. Reexamination of the role of the amino terminus of SecA in promoting its dimerization and functional state. J Bacteriol. 2008;190(21):7302–7. PubMed Central PMCID: PMC2580686. pmid:18723626
  40. 40. Papanikolau Y, Papadovasilaki M, Ravelli RB, McCarthy AA, Cusack S, Economou A, et al. Structure of dimeric SecA, the Escherichia coli preprotein translocase motor. Journal of molecular biology. 2007;366(5):1545–57. pmid:17229438
  41. 41. Patel CN, Smith VF, Randall LL. Characterization of three areas of interactions stabilizing complexes between SecA and SecB, two proteins involved in protein export. Protein science: a publication of the Protein Society. 2006;15(6):1379–86. PubMed Central PMCID: PMC2265093.
  42. 42. Tai PC, Tian G, Xu H, Lian JP, Yu JN. In vitro protein translocation into Escherichia coli inverted membrane vesicles. Methods Cell Biol. 1991;34:167–87. pmid:1943799
  43. 43. Dalal K, Duong F. The SecY complex forms a channel capable of ionic discrimination. EMBO Rep. 2009;10(7):762–8. PubMed Central PMCID: PMC2727429. pmid:19483671
  44. 44. Duong F, Wickner W. The SecDFyajC domain of preprotein translocase controls preprotein movement by regulating SecA membrane cycling. The EMBO journal. 1997;16(16):4871–9. PubMed Central PMCID: PMC1170122. pmid:9305629
  45. 45. Collinson I, Breyton C, Duong F, Tziatzios C, Schubert D, Or E, et al. Projection structure and oligomeric properties of a bacterial core protein translocase. The EMBO journal. 2001;20(10):2462–71. Epub 2001/05/15. PubMed Central PMCID: PMC125464. pmid:11350935
  46. 46. Huang YJ, Wang H, Gao FB, Li M, Yang H, Wang B, et al. Fluorescein analogues inhibit SecA ATPase: the first sub-micromolar inhibitor of bacterial protein translocation. ChemMedChem. 2012;7(4):571–7. Epub 2012/02/23. PubMed Central PMCID: PMC3483595. pmid:22354575
  47. 47. Lin BR, Gierasch LM, Jiang C, Tai PC. Electrophysiological studies in Xenopus oocytes for the opening of Escherichia coli SecA-dependent protein-conducting channels. The Journal of membrane biology. 2006;214(2):103–13. Epub 2007/05/29. PubMed Central PMCID: PMC2896742. pmid:17530158
  48. 48. Hsieh YH, Zou J, Jin JS, Yang H, Chen Y, Jiang C, et al. Monitoring channel activities of proteoliposomes with SecA and Cx26 gap junction in single oocytes. Analytical biochemistry. 2015;480:58–66. pmid:25862083
  49. 49. Lin BR, Hsieh YH, Jiang C, Tai PC. Escherichia coli membranes depleted of SecYEG elicit SecA-dependent ion-channel activity but lose signal peptide specificity. J Membr Biol. 2012;245(11):747–57. Epub 2012/08/03. pmid:22854753
  50. 50. Lill R, Dowhan W, Wickner W. The ATPase activity of SecA is regulated by acidic phospholipids, SecY, and the leader and mature domains of precursor proteins. Cell. 1990;60(2):271–80. pmid:2153463
  51. 51. Hu HJ, Holley J, He J, Harrison RW, Yang H, Tai PC, et al. To be or not to be: predicting soluble SecAs as membrane proteins. IEEE transactions on nanobioscience. 2007;6(2):168–79. Epub 2007/08/19. pmid:17695753
  52. 52. Floyd JH, You Z, Hsieh YH, Ma Y, Yang H, Tai PC. The dispensability and requirement of SecA N-terminal aminoacyl residues for complementation, membrane binding, lipid-specific domains and channel activities. Biochem Biophys Res Commun. 2014;453(1):138–42. PubMed Central PMCID: PMC4250336. pmid:25264203
  53. 53. Nakatogawa H, Mori H, Ito K. Two Independent Mechanisms Down-regulate the Intrinsic SecA ATPase Activity. Journal of Biological Chemistry. 2000;275(43):33209–12. pmid:10982786
  54. 54. Breukink E, Demel RA, de Korte-Kool G, de Kruijff B. SecA insertion into phospholipids is stimulated by negatively charged lipids and inhibited by ATP: a monolayer study. Biochemistry. 1992;31(4):1119–24. pmid:1531180
  55. 55. Breukink E, Nouwen N, van Raalte A, Mizushima S, Tommassen J, de Kruijff B. The C terminus of SecA is involved in both lipid binding and SecB binding. J Biol Chem. 1995;270(14):7902–7. pmid:7713885
  56. 56. Jilaveanu LB, Oliver D. SecA Dimer Cross-Linked at Its Subunit Interface Is Functional for Protein Translocation. Journal of Bacteriology. 2006;188(1):335–8. pmid:16352850
  57. 57. van Voorst F, van der Does C, Brunner J, Driessen AJ, de Kruijff B. Translocase-bound SecA is largely shielded from the phospholipid acyl chains. Biochemistry. 1998;37(35):12261–8. Epub 1998/09/02. pmid:9724540
  58. 58. Ding H, Mukerji I, Oliver D. Lipid and signal peptide-induced conformational changes within the C-domain of Escherichia coli SecA protein. Biochemistry. 2001;40(6):1835–43. pmid:11327846
  59. 59. Ding H, Mukerji I, Oliver D. Nucleotide and phospholipid-dependent control of PPXD and C-domain association for SecA ATPase. Biochemistry. 2003;42(46):13468–75. pmid:14621992
  60. 60. Keller RC. The prediction of novel multiple lipid-binding regions in protein translocation motor proteins: A possible general feature. Cell Mol Biol Lett. 2011;16(1):40–54. Epub 2010/10/20. pmid:20957445
  61. 61. Hartl F-U, Lecker S, Schiebel E, Hendrick JP, Wickner W. The binding cascade of SecB to SecA to SecY E mediates preprotein targeting to the E. coli plasma membrane. Cell. 1990;63(2):269–79. pmid:2170023
  62. 62. Mao C, Cheadle CE, Hardy SJ, Lilly AA, Suo Y, Sanganna Gari RR, et al. Stoichiometry of SecYEG in the active translocase of Escherichia coli varies with precursor species. Proc Natl Acad Sci U S A. 2013;110(29):11815–20. PubMed Central PMCID: PMC3718118. pmid:23818593
  63. 63. Cooper DB, Smith VF, Crane JM, Roth HC, Lilly AA, Randall LL. SecA, the motor of the secretion machine, binds diverse partners on one interactive surface. Journal of molecular biology. 2008;382(1):74–87. PubMed Central PMCID: PMC2633600. pmid:18602400
  64. 64. Das S, Oliver DB. Mapping of the SecA.SecY and SecA.SecG interfaces by site-directed in vivo photocross-linking. J Biol Chem. 2011;286(14):12371–80. PubMed Central PMCID: PMC3069440. pmid:21317284
  65. 65. Snyders S, Ramamurthy V, Oliver D. Identification of a region of interaction between Escherichia coli SecA and SecY proteins. J Biol Chem. 1997;272(17):11302–6. pmid:9111035
  66. 66. Tsukazaki T, Mori H, Echizen Y, Ishitani R, Fukai S, Tanaka T, et al. Structure and function of a membrane component SecDF that enhances protein export. Nature. 2011;474(7350):235–8. PubMed Central PMCID: PMC3697915. pmid:21562494
  67. 67. Arkowitz RA, Wickner W. SecD and SecF are required for the proton electrochemical gradient stimulation of preprotein translocation. The EMBO journal. 1994;13(4):954–63. PubMed Central PMCID: PMC394897. pmid:8112309
  68. 68. Duong F, Wickner W. Distinct catalytic roles of the SecYE, SecG and SecDFyajC subunits of preprotein translocase holoenzyme. The EMBO journal. 1997;16(10):2756–68. PubMed Central PMCID: PMC1169885. pmid:9184221
  69. 69. Pogliano JA, Beckwith J. SecD and SecF facilitate protein export in Escherichia coli. The EMBO journal. 1994;13(3):554–61. PubMed Central PMCID: PMC394844. pmid:8313900
  70. 70. Jilaveanu LB, Zito CR, Oliver D. Dimeric SecA is essential for protein translocation. Proceedings of the National Academy of Sciences of the United States of America. 2005;102(21):7511–6. pmid:15897468
  71. 71. Wang H, Ma Y, Hsieh YH, Yang H, Li M, Wang B, et al. SecAAA trimer is fully functional as SecAA dimer in the membrane: existence of higher oligomers? Biochem Biophys Res Commun. 2014;447(2):250–4. PubMed Central PMCID: PMC4036620. pmid:24704204
  72. 72. Yang CK, Ewis HE, Zhang X, Lu CD, Hu HJ, Pan Y, et al. Nonclassical protein secretion by Bacillus subtilis in the stationary phase is not due to cell lysis. J Bacteriol. 2011;193(20):5607–15. Epub 2011/08/23. PubMed Central PMCID: PMC3187209. pmid:21856851
  73. 73. Duong F, Wickner W. The PrlA and PrlG phenotypes are caused by a loosened association among the translocase SecYEG subunits. The EMBO journal. 1999;18(12):3263–70. PubMed Central PMCID: PMC1171407. pmid:10369667
  74. 74. Neumann-Haefelin C, Schäfer U, Müller M, Koch H-G. SRP-dependent co-translational targeting and SecA-dependent translocation analyzed as individual steps in the export of a bacterial protein. The EMBO journal. 2000;19(23):6419–26. pmid:11101515
  75. 75. Qi H-Y, Bernstein HD. SecA Is Required for the Insertion of Inner Membrane Proteins Targeted by the Escherichia coli Signal Recognition Particle. Journal of Biological Chemistry. 1999;274(13):8993–7. pmid:10085146
  76. 76. Butkus ME, Prundeanu LB, Oliver DB. Translocon "pulling" of nascent SecM controls the duration of its translational pause and secretion-responsive secA regulation. J Bacteriol. 2003;185(22):6719–22. PubMed Central PMCID: PMC262105. pmid:14594848
  77. 77. Kramer G, Boehringer D, Ban N, Bukau B. The ribosome as a platform for co-translational processing, folding and targeting of newly synthesized proteins. Nat Struct Mol Biol. 2009;16(6):589–97. pmid:19491936
  78. 78. Singh R, Kraft C, Jaiswal R, Sejwal K, Kasaragod VB, Kuper J, et al. Cryo-electron microscopic structure of SecA protein bound to the 70S ribosome. J Biol Chem. 2014;289(10):7190–9. PubMed Central PMCID: PMC3945378. pmid:24443566
  79. 79. Wu ZC, de Keyzer J, Kedrov A, Driessen AJ. Competitive binding of the SecA ATPase and ribosomes to the SecYEG translocon. J Biol Chem. 2012;287(11):7885–95. PubMed Central PMCID: PMC3318685. pmid:22267723