ACEMBLing a Multiprotein Transmembrane Complex

CHAPTER TWO

ACEMBLing a Multiprotein Transmembrane Complex: The Functional SecYEG-SecDF-YajCYidC Holotranslocon Protein Secretase/Insertase Joanna Komar*,1, Mathieu Botte†,{,1, Ian Collinson*, Christiane Schaffitzel*,†,{, Imre Berger*,†,{,2 *School of Biochemistry, University of Bristol, Bristol, United Kingdom † European Molecular Biology Laboratory, Grenoble, France { Unit for Virus Host-Cell Interactions, University of Grenoble Alpes-EMBL-CNRS, Unite´ mixte de Recherche, Grenoble, France 1 Equal contribution. 2 Corresponding author: e-mail address: [email protected]

Contents 1. 2. 3. 4.

Introduction ACEMBLing the HTL Multiprotein Complex Purifying the HTL HTL Integrity and Activity 4.1 Incorporation of translocation complexes in proteoliposomes 4.2 Orientation of the reconstituted complexes 4.3 Subunit interactions and activity of reconstituted translocation complexes 5. Discussion and Conclusions Acknowledgments References

24 27 28 32 32 34 36 43 46 46

Abstract Membrane proteins constitute about one third of the proteome. The ubiquitous Sec machinery facilitates protein movement across or integration of proteins into the cytoplasmic membrane. In Escherichia coli post- and co-translational targeting pathways converge at the protein-conducting channel, consisting of a central pore, SecYEG, which can recruit accessory domains SecDF-YajC and YidC, to form the holotranslocon (HTL) supercomplex. Detailed analysis of HTL function and architecture remained elusive until recently, largely due to the lack of a purified, recombinant complex. ACEMBL is an advanced DNA recombineering-based expression vector system we developed for producing challenging multiprotein complexes. ACEMBL affords the means to combine multiple expression elements including promoter DNAs, tags, genes Methods in Enzymology, Volume 556 ISSN 0076-6879 http://dx.doi.org/10.1016/bs.mie.2014.12.027

#

2015 Elsevier Inc. All rights reserved.

23

24

Joanna Komar et al.

of interest, and terminators in a combinatorial manner until optimal multigene expression plasmids are constructed that yield correctly assembled, homogenous, and active multiprotein complex specimens. We utilized ACEMBL for recombinant HTL overproduction. We developed protocols for detergent solubilizing and purifying the HTL. Highly purified complex was then used to reveal HTL function and the interactions between its constituents. HTL activity in protein secretion and membrane protein insertion was analyzed in both the presence and absence of the proton-motive force. Setting up ACEMBL for the assembly of multigene expression constructs that achieve high yields of functional multisubunit membrane protein complex is straightforward. Here, we used ACEMBL for obtaining active HTL supercomplex in high quality and quantity. The concept can likewise be applied to obtain many other assemblies of similar complexity, by overexpression in prokaryotic, and also eukaryotic hosts.

1. INTRODUCTION In vivo, the bacterial core translocon SecYEG interacts with additional components: YidC and SecDF-YajC, which together form a complex referred to as the holotranslocon (HTL). The accessory subunits are believed to stimulate both post- and co-translational translocation (Arkowitz & Wickner, 1994; Beck et al., 2001; Duong & Wickner, 1997b; Luirink, von Heijne, Houben, & de Gier, 2005; Scotti et al., 2000; Urbanus et al., 2001); however, their mechanism of action remains largely unknown. The majority of research in the Sec translocation field has focused on the core translocon SecYEG. This heterotrimeric protein-conducting channel has been extensively characterized over the past 20 years, and one of the major advances was the appearance of its first X-ray structure a decade ago (van den Berg et al., 2004). However, the understanding of the role of individual components of the HTL as part of the complex still remains poor. The main approaches taken to investigate the interactions between different subunits within the HTL involve co-immunoprecipitation and cross-linking studies. Interactions involved in the formation of the HTL complex were first reported by Duong and Wickner, who coimmunoprecipitated SecYEG together with SecDF and YajC from digitonin-solubilized membranes (Duong & Wickner, 1997a). It was later discovered that the back then newly characterized membrane protein YidC also copurifies with SecYEG (Scotti et al., 2000). In this study, overexpression of translocase complexes, SecYEG and SecDF-YajC, resulted in a significant elevation of YidC levels, suggesting that YidC is also part of the HTL complex. Moreover, YidC and SecYEG were also shown to

ACEMBLing Multisubunit Membrane Protein Complexes

25

copurify during membrane protein insertion (Boy & Koch, 2009). Depletion studies revealed another functional interaction within the HTL complex between SecDF-YajC and one of the subunits of the SecYEG channel, SecG (Kato, Nishiyama, & Tokuda, 2003). More recently, a thorough cross-linking analysis reported that YidC makes contacts with the lateral gate of SecYEG channel. The interaction is seemingly dynamic upon ribosome and ribosome nascent chain binding (Sachelaru et al., 2013). This study also confirmed interactions between YidC and SecDF-YajC (YidC–SecF interaction was previously reported in Xie, Kiefer, Nagler, Dalbey, & Kuhn, 2006), which however appear nonessential for the YidC-SecY contacts (Sachelaru et al., 2013). X-ray crystal structures of both SecDF (Tsukazaki et al., 2011) and YidC (Kumazaki et al., 2014) have been solved recently. This structural information has shed light onto the function of these complexes and created a new scope for their analysis. Nevertheless, it is difficult to rationalize the function of the individual structures in isolation; therefore, their role in the context of their physiological partners, as part of the HTL, still remains largely unclear. A thorough biochemical, structural, and functional analysis of the HTL has not been possible yet, due to the lack of an isolated stable complex containing a full complement of its seven subunits. Moreover, the heterogeneous character of the complex and its low copy number in cells render its isolation from native source material exceedingly difficult. In order to overcome this challenge, attempts have been previously made to coreconstitute individual subunits of the HTL together into liposomes after their independent purifications. This strategy, however, has its limitations, as individual subunits might not be in the correct orientation competent for complex formation outside their native membrane environment. Therefore, until recently, efforts in studying the architecture and function of this multiprotein complex assembly have remained unsuccessful. Purification of protein complexes, and in particular of membrane protein complexes, in the quality and quantity required for detailed molecular level functional analysis remains a considerable challenge to date. Most complexes in cells do not exist in sufficient abundance to allow efficient extraction from native source material. Furthermore, endogenous material is often characterized by considerable heterogeneity, related not only to functional aspects but also owing to the fact that the cell at a given time will be dynamically assembling and degrading the complexes that catalyze cellular function. Recombinant overexpression has made a tremendous impact on studying proteins and their interactions, and heterologous production has largely

26

Joanna Komar et al.

replaced purification of endogenous material in molecular biology laboratories. Much effort has been expended to develop expression systems that can produce not only individual proteins but also multisubunit complexes, into which most of the proteins necessarily assemble to catalyze cellular function. Powerful expression systems for prokaryotic and eukaryotic hosts are being developed and implemented. For instance, the MultiBac system we developed has been particularly useful to produce high-quality multisubunit complexes that could not be produced in Escherichia coli but rely on eukaryotic expression (Barford, Takagi, Schultz, & Berger, 2013; Bieniossek, Imasaki, Takagi, & Berger, 2012; Fitzgerald et al., 2006). A recent development in expression and purification of multiprotein complexes (Bieniossek et al., 2009) has presented a new opportunity to construct a single multigene expression plasmid encoding all seven subunits of the HTL. This innovative technology, ACEMBL, constitutes a rapid and versatile tool for production of stable multisubunit complexes in E. coli in high quality and quantity and has been successfully used for expression of a range of samples including multiprotein complexes and nucleic acid/ protein assemblies (Bieniossek et al., 2009). Often, balancing expression levels and placing purification tags are critical issues for successful heterologous production of complexes. The ACEMBL concept relies on using DNA assembly strategies that allow to combinatorially permutate the elements required for expressing genes coding for individual components of a complex (Bieniossek et al., 2009). These elements include weak and strong promoters to drive transcription, genes of interest, purification tags, proteolytic cleavage sites, and terminator elements, giving rise to expression cassettes that will lead to producing each subunit of a protein complex at a defined level and in the defined setup, in the coexpression experiment. ACEMBL likewise provides the means to combine several genes into polycistronic cassettes and, in addition, to combine several expression cassettes rapidly into multigene expression cassettes (Bieniossek et al., 2009). The underlying technology was termed “tandem recombineering” (TR) and involves a combination of sequence- and ligation-independent multifragment DNA assembly (Li & Elledge, 2007) in conjunction with DNA fusion catalyzed by the Cre recombinase. Cre is a site-specific recombination enzyme, which combines DNAs containing a specific repeat recognition sequence, LoxP (Bieniossek et al., 2009; Fitzgerald et al., 2006). ACEMBL also comprises an array of custom-designed synthetic plasmid modules called acceptors and donors, which can be conjoined by means of Cre-catalyzed plasmid fusion. ACEMBL was originally developed to enable structural genomics

ACEMBLing Multisubunit Membrane Protein Complexes

27

pipelines to perform protein complex structure determination in high throughput, in a parallelized setup relying on robotics (Bieniossek et al., 2009; Trowitzsch, Bieniossek, Nie, Garzoni, & Berger, 2010; Vijayachandran et al., 2011). However, all steps involved can likewise be performed in manual mode, and the protocols developed for robots are also highly efficient when used manually. The ACEMBL system was first used to produce a heterohexameric complex SecYEG-SecDF-YidC, consisting of 33 unique transmembrane segments (TMSs), in E. coli as an expression host (Bieniossek et al., 2009). However, as soon as it became clear that YajC is also an integral part of the HTL, we made use of the flexibility of the ACEMBL system, allowing easy modification of multigene constructs, to integrate an additional expression module for producing YajC. This enabled us to successfully produce and purify the complete HTL supercomplex (Schulze et al., 2014). Being able to purify a stable complex containing a full complement of all seven HTL subunits allowed subsequent analysis of its organization and function. For this purpose, protein secretion and membrane protein insertion processes were reconstituted from purified components in vitro, which then provided efficient means to elucidate the activity of the HTL in both post- and co-translational translocation, providing unique insight into the function of this vital multiprotein transmembrane machine.

2. ACEMBLing THE HTL MULTIPROTEIN COMPLEX The ACEMBL system was developed to tackle the challenge of producing multiprotein complexes, in particular transmembrane protein complexes for structural and functional studies (Bieniossek et al., 2009). Because ACEMBL was originally designed as part of a structural genomics pipeline, the protocols and procedures were optimized to be sufficiently robust to function in a robotized, high-throughput environment (Bieniossek et al., 2009; Vijayachandran et al., 2011). Automation requires simple procedures that can be carried out reliably in parallel and many times, with very high accuracy and consistency. The ACEMBL system meets these criteria by relying on an array of custom-designed plasmid DNA reagents, comprising acceptor and donor plasmids, and a procedure called TR. TR is used to create individual expression cassettes containing single genes or polycistrones for insertion into individual acceptor and donor plasmids, which are then combined into functional multigene expression constructs by means of plasmid fusion catalyzed by the Cre enzyme. TR consists of a sequence- and

28

Joanna Komar et al.

ligation-independent DNA assembly step to combine DNA that represent the individual elements of a functional expression cassette, and the Cre– LoxP fusion to combine the plasmid modules fitted with such expression cassettes into multigene constructs (Haffke, Viola, Nie, & Berger, 2013). Cre is a site-specific recombinase, which recognizes and recombines specific sequence elements representing an imperfect inverted repeat. All acceptors and donors of the ACEMBL suite are equipped with this so-called LoxP DNA sequence and as a consequence can be fused by Cre recombinase. Each donor and acceptor can be fitted with one or several expression cassettes, each comprising single cistrones or polycistrones, and combined into acceptor–donor fusions by using TR. The ACEMBL system is illustrated schematically in Fig. 1. Overexpression of protein complexes may require particular attention with respect to the production of the individual subunits of the complex studied. One or several subunits may be badly produced in the coexpression experiment, thus limiting overall yield and potentially giving rise to heterogeneity of the specimen produced. ACEMBL provides the means to balance expression levels of subunits by rapidly testing in a combinatorial manner a number of promoter systems (T7, trc, lac and arabinose) to regulate transcription rates and consequently translation of subunits. This asset turned out to be particularly useful for HTL supercomplex production. A series of promoters and gene combinations into polycistrones were tested in parallel by using ACEMBL, along with varied placement of purification tags, ultimately resulting in a series of pACEMBL-HTL expression constructs, which could be used for efficient and balanced production of the HTL and its constituent subcomplexes (Fig. 2).

3. PURIFYING THE HTL The pACEMBL-HTL multigene expression plasmids, generated by the TR technique and ACEMBL reagents, were transformed into E. coli C43(DE3) cells as a host to minimize potential toxicity associated with the overexpression of membrane proteins (Miroux & Walker, 1996). Freshly transformed E. coli cells were grown in 2
YT broth with antibiotics to an OD600 of 0.8. Induction using 1 mM IPTG and 0.2% (w/v) arabinose was followed by further 3 h of incubation in shaker flasks. Denaturing polyacrylamide gel electrophoresis (SDS-PAGE) analysis confirmed balanced overexpression of all HTL subunits. Cell pellets were harvested by centrifugation and stored following common procedures.

29

ACEMBLing Multisubunit Membrane Protein Complexes

Acceptor T7

MIE

T7

LoxP

ADD

MIE

LoxP BR322 R Ap ori

BR322 ori

pACE

R Tc

pACE-DS-DK

Cre

pACE2

Donor T7

pACE-DK-DS

MIE

pDC

LoxP R R6Kg Kn ori

pDK

DNA assembly (SLIC, conventional)

AD

Lac MIE

Lac MIE LoxP

R R6Kg Cm ori

De-Cre

LoxP

pACE-DK

R6Kg R Sp ori

pDS

LoxP R R6Kg Kn ori

pDK

Multigene fusion (Cre-LoxP)

pACE-DS LoxP LoxP

R R6Kg Sp ori

pDS

BR322 R Ap ori

pACE

Complex production

Figure 1 The ACEMBL system for protein complex production. The ACEMBL system is depicted in a schematic view. ACEMBL comprises an array of plasmid modules (top-left). These plasmid modules were custom designed and contain exclusively DNA elements with defined functions. Acceptors contain a common origin of replication (derived from plasmid BR322); donors contain a conditional origin of replication (derived from R6Kγ phage). Donors can exist in regular cloning strains only if fused to an acceptor (Bieniossek et al., 2009; Haffke et al., 2013). Donors and acceptors are fitted with a blank expression cassette containing a multiple insertion element (MIE) for pasting in DNA elements (genes, tags, and proteolytic sites) of interest. All donors and acceptors contain a LoxP imperfect inverted sequence for Cre-mediated plasmid fusion (shown on the right). Expression cassettes driven by T7 and lac promoters are shown; however, any other promoter (such as ara, trc, and others) can be also placed in the plasmid modules. Ap, ampicillin; Tc, teracyclin; Cm, chloramphenicol; Kn, kanamycin; Sp, spectinomycin. Cre-mediated plasmid fusion is depicted schematically on the right. The Cre reaction is an equilibrium reaction, consisting of fusion (arrow left) and concomitant excision (arrow right). Incubation of donors and an acceptor with Cre therefore generates a variety of fusions in a single reaction. These are all characterized by their resistance marker combination and can thus be selected by antibiotics challenge (Bieniossek et al., 2009; Haffke et al., 2013). The combination of DNA element assembly by sequence- and ligation-independent cloning (SLIC) methods, and Cre-LoxP-mediated plasmid fusion is called tandem recombineering (TR). Multigene expression plasmids combinatorially generated are then used for protein complex production. The ACEMBL system shown here that was used for producing the HTL is designed for expression in a prokaryotic host. Variants of ACEMBL with mammalian or baculoviral promoters were likewise developed, if expression in eukaryotic hosts is to be carried out (Vijayachandran et al., 2011). Modified from Bieniossek et al. (2009).

30

Joanna Komar et al.

Figure 2 Holotranslocon expression constructs based on the ACEMBL expression system. HTL1 (top-left) consists of the pACE acceptor and the pDC donor vector combined by Cre-loxP fusion (LoxP, gray circles). A polycistron encoding for YidC, SecD (D), SecF (F) with an arabinose promoter (ara, yellow) has been subcloned into pACE. A second polycistron encoding for SecY, SecE (E), and SecG (G) with a trc promoter (trc, red) has been cloned into pDC. HTL2 (middle-left) is a fusion of the HTL1 construct with the donor vector pDK encoding Strep-tagged YajC with lactose promoter (lac, yellow). HTL3 (bottom-left) is a fusion of HTL1 with pDK encoding CBP-tagged YajC with trc promoter. HTL4 (top-right) is a HTL3-based construct with a Strep-tagged YidC and a tagfree SecD, where all genes are under the control of a trc promoter. DFYY (bottom-right) is a fusion of the previous pACE and pDK vectors. The position of hexahistidine tags in YidC, SecD, and SecE is indicated in green. The position of the CBP-tag in YajC is indicated in gray. The transcription terminators are shown as black squares. Origins of replication (BR322 and R6Kγori) are indicated. Antibiotic resistance genes confer resistance to the following antibiotics: Ap, ampicillin; Cm, chloramphenicol; and Kn, kanamycin.

Subsequently, cell pellets were broken at 25 kpsi using a cell disruptor (Constant Systems, Ltd., Daventry, UK) in TSG130 buffer (20 mM Tris– Cl, pH 8.0, 130 mM NaCl, 10% (v/v) glycerol). Next, the membranes were collected and solubilized by rotation in TSG130 buffer containing 2% (w/v) n-dodecyl-β-D-maltoside (DDM), incubated for 1 h at a temperature of 4 °C. The DDM-soluble fraction was clarified by further centrifugation. The complete HTL complex was purified by metal affinity chromatography followed by size-exclusion chromatography, combined with an anion exchange step. To this end, the cleared DDM-soluble fraction was then applied to metal affinity purification using a chelating Ni2+-Sepharose Fast Flow column (GE Healthcare), which was preequilibrated with TSG130-

31

ACEMBLing Multisubunit Membrane Protein Complexes

DDM buffer that contained 0.1% DDM. After washing the resin bed with 10 column volumes of TSG130-DDM wash buffer containing 30 mM imidazole, HTL supercomplex was eluted with TSG130-DDM elution buffer containing 500 mM imidazole. Peak fractions were collected and pooled, and then applied to a Superdex 200, 26/60 gel filtration column (GE Healthcare). A Q-Sepharose ion exchange column equilibrated in TSG130 + 0.05% DDM was placed in line with the Superdex 200 column. A well-defined A280 peak eluted at an elution volume of approximately 190 ml. Analysis of the adjacent fractions across the peak by SDS-PAGE demonstrated the comigration of all subunits within a single complex (Schulze et al., 2014; Fig. 3). The HTL supercomplex was applied to a 50-kDa-molecular weight (MW) cut-off centrifugation filter (Amicon) and concentrated to 10 mg ml-1, using an experimentally determined molar extinction coefficient of εHTL ¼ 497,000 M-1 cm-1. The HTL was stored at 4 °C or frozen in glycerol at 80 °C.

A

B

ara

ori

YidC

HTL3

Accessory domains

SecD

SecF

D trc

R BR322 Ap ori

YajC

R Cm

SecY

trc

YajC-CBP 16 18 20 22 24

300

SecG

Protein-conducting channel

A280 (mAU)

SecE

R Kn

SecY E G

400

YidC

R6Kg

F

60 50 40

SecD-His YidC-His

30

SecY SecF

20

200

SecE-His SecG YajC-CBP

15 10

100 SDS-PAGE

100

200

300

400

Elution volume (ml)

Figure 3 Highly purified recombinant HTL. (A) Complete functional HTL holocomplex comprises the protein-conducting channel SecYEG and the accessory domains formed by SecD, SecF, YidC, and YajC. (B) Construct pACEMBL-HTL (top) was used for carefully balanced coexpression of the HTL subunits, resulting in efficient complex assembly in the expression host. Purification involving metal affinity (IMAC), size-exclusion (SEC), and ion exchange (IEX) chromatography results in highly purified HTL, which eluted in SEC as a single peak containing all subunits. Modified from Schulze et al. (2014).

32

Joanna Komar et al.

4. HTL INTEGRITY AND ACTIVITY 4.1 Incorporation of translocation complexes in proteoliposomes With the aim to study protein secretion and membrane protein insertion activities of the HTL, compared to the core translocon SecYEG, which was purified according to published protocols (Gold et al., 2010), purified proteins were first incorporated into phospholipid vesicles to form proteoliposomes (PLs), both in the presence and absence of bacteriorhodopsin (BR), as described previously (Collinson et al., 2001; Schulze et al., 2014). SecYEG has been previously shown to form dimers in the membrane, which can dissociate into monomers in the presence of increasing detergent concentrations (Bessonneau, Besson, Collinson, & Duong, 2002). Therefore, both purified complexes, SecYEG and the HTL, were first analyzed by Blue Native-PAGE in order to determine their oligomeric states (Schulze et al., 2014). Both complexes dissociated into their subcomplexes when subjected to increasing detergent concentrations, generating two bands migrating at 300 and 150–200 kDa. The apparent MW of the intact HTL complex was about the same as that of the SecYEG dimer (300 kDa), which suggests that the HTL consists of single copies of SecYEG and SecDF-YajC-YidC. The lower band of the dissociated HTL had the same apparent MW (150–200 kDa) as that of SecYEG monomers formed after dissociation of SecYEG dimers, as well as that of the SecDF-YidC complex. Analysis of this lower band by second dimension SDS-PAGE revealed presence of all components of the HTL, suggesting separation of the complex into SecYEG and SecDF-YajC-YidC (Schulze et al., 2014). The behavior of the HTL complex resembles that of SecYEG dimers dissociating into monomers (Bessonneau et al., 2002) and could be a result of high detergent concentrations causing the extraction of tightly bound lipids within the complex (Gold et al., 2010). Based on these findings, proteins were reconstituted assuming a homodimeric form of SecYEG and a heterodimeric complex of the HTL, consisting of one copy of each subcomplex, SecYEG and SecDF-YajC-YidC. Resulting PLs were subsequently analyzed by SDS-PAGE electrophoresis followed by Coomassie blue staining (Fig. 4), in order to assess efficiency of the reconstitution.

ACEMBLing Multisubunit Membrane Protein Complexes

33

Figure 4 Analysis of SecYEG and HTL incorporated into proteoliposomes. Aliquots of vesicles containing SecYEG and HTL in the absence (A) and presence (B) of bacteriorhodopsin (BR) were solubilized in LDS sample loading buffer and analyzed by SDS-PAGE electrophoresis followed by Coomassie blue staining. Modified from Schulze et al. (2014).

As expected from the reconstitution stoichiometries, the quantity of SecY was higher in the SecYEG vesicles compared to the HTL vesicles, regardless of the presence of BR (Fig. 4). This was due to the presence of two SecY copies in the SecYEG homodimer compared to only one SecY copy in the HTL heterodimer, corresponding to an intended equal concentration of homo- and heterodimers. The relative amounts of BR were similar in all sets of BR-containing vesicles. The noticeable slightly higher migration of SecE in the HTL samples compared to SecYEG samples is a result of a minor sequence variation in their affinity tags. Subsequent analysis of the PLs by Blue Native-PAGE (Fig. 5) revealed the same dissociation pattern in the presence of increasing detergent concentrations as previously observed with purified complexes. A covalently linked SecY dimer (Y–Y) was used as a MW reference. Generated bands corresponded to 350 and 200 kDa, migrating slightly higher compared to purified proteins due to the presence of lipids. The apparent MW of the intact HTL heterodimer was about the same as that of the SecYEG dimer (350 kDa), as observed with purified proteins. The lower band, representing the separated HTL complex into SecYEG

34

Joanna Komar et al.

Figure 5 Analysis of HTL- and SecYEG-containing proteoliposomes by Blue Native (BN)PAGE. Twenty-five nanogram of complexes reconstituted into lipid vesicles were incubated with decreasing detergent concentrations (from the left: 0.1%, 0.05%, and 0.02% DDM) for 30 min at 4 °C. Samples were subsequently analyzed by BN-PAGE followed by silver staining. A covalently linked SecY–Y dimer was used as a molecular weight reference.

and SecDF-YajC-YidC subcomplexes, had the same apparent MW as that of SecYEG monomers formed after SecYEG dimer dissociation.

4.2 Orientation of the reconstituted complexes In order to analyze the orientation of the reconstituted complexes in PLs, i.e., what proportion of complexes was in the secretion- and insertioncompetent inverted orientation (with cytosolic functional sites facing outward, exposed for ribosome and SecA binding; Fig. 6), SecYEG- and HTL-containing PLs were subjected to trypsin proteolysis followed by SDS-PAGE analysis and Coomassie staining (Fig. 7). Treatment of SecYEG-containing PLs with trypsin resulted in the generation of a 21-kDa band corresponding to the N-terminal fragment of SecY (Akiyama & Ito, 1990; Brundage, Hendrick, Schiebel, Driessen, & Wickner, 1990) due to cleavage between TMSs 6 and 7 (Robson, Booth, Gold, Clarke, & Collinson, 2007; Fig. 7). This indicates that the majority of reconstituted protein favored an outwardly facing orientation of its cytosolic sites. Similarly, subjecting the HTL-containing PLs to trypsin treatment resulted in the expected size shift for the SecY band (Fig. 7). This sensitivity

Figure 6 Inverted orientation of E. coli SecY and YidC in proteoliposomes. The presented orientation corresponds to the cytoplasmic side of the proteins being exposed to the outside of the vesicles. A trypsin cleavage site in SecY is indicated by a pair of scissors.

Figure 7 Orientation of translocation complexes in proteoliposomes. SecYEG- and HTLcontaining vesicles were subjected to trypsin proteolysis and subsequently analyzed by SDS-PAGE electrophoresis followed by Coomassie blue staining. Generated proteolysis fragments are indicated by a black arrowhead. Modified from Schulze et al. (2014).

36

Joanna Komar et al.

of the SecY subunit to proteolysis indicates that its cytosolic side is facing the outside of the vesicles as part of the HTL complex. Taken together, proteolysis of the HTL-containing PLs suggests that the cytoplasmic cleavage site between TMSs 6 and 7 in SecY is mainly exposed to the outside of the vesicles, as is the case for SecYEG-containing PLs. This suggests that the reconstitution process favored the orientation of the proteins with their cytosolic functional sites facing outward in PLs, exposed for interactions with their binding partners.

4.3 Subunit interactions and activity of reconstituted translocation complexes Cross-linking of inner membrane vesicles (IMVs) containing either overexpressed SecYEG or the HTL generates different cross-linking patterns (Schulze et al., 2014). This indicates the characteristic nature of subunit interactions within those complexes. As cross-linking products’ characteristics of the SecYEG dimer were either missing or reduced in the HTL sample, this suggests that there is only one copy of SecYEG present in the HTL complex. In order to confirm that these interactions were maintained after reconstitution of the complexes into phospholipid vesicles and that this process did not affect integrity of the HTL, SecYEG- and HTLcontaining PLs were subjected to nonspecific photo-cross-linking using Tris-bipyridylruthenium(II) (PICUP), as described previously (Deville et al., 2011; Fancy & Kodadek, 1999). After irradiation by visible light, samples were analyzed by SDS-PAGE electrophoresis followed by Coomassie staining (Fig. 8A) or Western blotting against SecY, SecE, and SecG (Fig. 8B). A covalently linked SecY dimer (Y–Y) was used as a MW reference for formed products. A similar pattern of cross-linking products was observed with HTL- and SecYEG-containing PLs compared to when IMVs overexpressing those complexes were used. However, cross-linking of the PL samples was not as efficient as that of IMVs, likely due to removing the complexes from their native membranes. SecY–Y and SecE–E contacts characteristic of SecYEG dimers were reduced in the PLs containing the HTL complex compared to those containing SecYEG. Moreover, cross-linking of the HTL sample resulted in detection of new higher MW complexes by the α-SecG antibody (highlighted in a red (gray color in the print version) box, Fig. 8B). Based on their MW, these products likely correspond to SecG–SecD or SecG–YidC contacts in the HTL complex. These products were also observed in cross-linked IMVs containing overexpressed HTL, and their

ACEMBLing Multisubunit Membrane Protein Complexes

37

Figure 8 Analysis of the subunit interactions within SecYEG and HTL by photoinduced cross-linking of unmodified proteins (PICUP). SecYEG- and HTL-containing proteoliposomes were irradiated for 10 s in the presence of either 0.8 or 2 mM Trisbipyridylruthenium(II) and subsequently analyzed by SDS-PAGE followed by Coomassie staining (A) or Western blotting (B). A covalently linked SecY–Y dimer was used as a molecular weight reference. Red (gray color in the print version) box indicates crosslinking products present only in the HTL sample but not in the SecYEG sample.

subsequent analysis by mass spectrometry revealed the presence of both YidC and SecD (Schulze et al., 2014). These results confirm that the HTL reconstituted into PLs consists of one copy of SecYEG and one copy of SecDF-YajC-YidC subcomplex, which replaces the second copy of SecYEG present in the SecYEG dimer. This is consistent with observations made when IMVs were subjected to the same analysis and shows that the subunit interactions and their organization within the HTL complex and the SecYEG dimer are maintained after incorporation into phospholipid vesicles. The presence of SecYEG core translocon in the HTL complex suggests that the HTL might be competent in protein secretion. Therefore, following successful reconstitution of the complex into lipid vesicles, its ability to translocate a secretory substrate proOmpA was tested. 4.3.1 ATP-stimulated protein secretion by SecA ATPase In order to test the ability of the HTL to translocate proOmpA into phospholipid vesicles, PLs containing the reconstituted complex were incubated in the presence of SecA, ATP, and the outer membrane precursor protein proOmpA in an in vitro translocation assay (Fig. 9). Protease-protected proOmpA successfully translocated inside the lumen of PLs was detected by Western blotting (Fig. 10). Interestingly, ATPdependent secretion activity of the HTL driven by SecA was severely reduced compared to SecYEG. No translocation was observed in the presence of empty vesicles or in the absence of the ATP, as expected.

38

Joanna Komar et al.

Figure 9 Translocation assay setup. Translocation complexes reconstituted into proteoliposomes are incubated in the presence of SecA, ATP, and a secretory substrate proOmpA. Following a 30-min incubation, reactions are treated with proteinase K in order to digest any substrate left on the outside of the vesicles. Protease-protected proOmpA that has been successfully translocated inside the lumen of the vesicles is detected by Western blotting.

Figure 10 In vitro translocation assay comparing secretion activity of SecYEG and the HTL complex. The graph shows percentage of proOmpA translocated inside the lumen of vesicles by SecYEG or the HTL complex. Empty vesicles () were used as a negative control. Quantification was performed using ImageJ software and values represent the mean of three independent experiments SEM.

ACEMBLing Multisubunit Membrane Protein Complexes

39

As demonstrated by cross-linking of the PLs used in the translocation assay above (Fig. 8), this reduced activity in proOmpA translocation was not a result of dissociation of the HTL into subcomplexes of SecYEG and SecDF-YajC-YidC, and subsequent SecYEG dimer formation. Interestingly, the translocation ATPase activity of SecA in the presence of either SecYEG- or HTL-containing PLs was very similar; however, the affinity of a translocation peptide for the HTL was slightly higher (Schulze et al., 2014). Therefore, the conversion of ATP to transport seems much lower for the HTL complex compared to SecYEG. 4.3.2 PMF-stimulated protein secretion activity Bacterial membranes can generate the proton-motive force (PMF) through the electron transport chain, and it is known that SecYEG transduces the energy available in the PMF to stimulate protein translocation (Brundage et al., 1990; Driessen, 1992; Geller, Movva, & Wickner, 1986; Schiebel, Driessen, Hartl, & Wickner, 1991; Shiozuka, Tani, Mizushima, & Tokuda, 1990). As both ATP and PMF are driving forces of protein translocation in vivo, acting at different stages of translocation (Schiebel et al., 1991), the ability of the HTL to translocate proOmpA in the presence of the PMF was tested. For that purpose, the light-driven proton pump BR was coreconstituted together with the translocation complexes into PLs in order to generate the PMF in the translocation assay setup (Fig. 11). These BR-containing PLs were used in order to study the PMFstimulated protein secretion mediated by the HTL compared to the SecYEG channel (Fig. 12). Translocation reactions were either illuminated to create the PMF (+Light) or kept in the dark in the absence of the PMF (Light). In order to collapse the PMF and therefore verify its existence, an uncoupling ionophore CCCP was added to the reactions in the presence of light (+CCCP). In the presence of the PMF (+Light), the efficiency of SecYEGmediated transport of proOmpA was doubled compared to when no PMF was present (Light; Fig. 12). When the uncoupling ionophore CCCP was added to the reaction in the presence of light, transport returned to levels present in the absence of light, verifying the collapse of an existing PMF. In the case of the HTL complex, the stimulation by the PMF was more impressive, resulting in a much larger increase in its transport efficiency (Fig. 12). However, the basal translocation activity of the HTL in the

40

Joanna Komar et al.

Figure 11 Proton-motive-force-stimulated translocation assay setup. Translocation complexes are coreconstituted into proteoliposomes together with bacteriorhodopsin (BR), which provides means of generating the proton-motive force in the system. Formed proteoliposomes are incubated in the presence of SecA, ATP, and a secretory substrate proOmpA for 30 min in front of a slide projector fitted with a yellow (light gray color in the print version) filter, or in the dark for reactions without the presence of the PMF, followed by proteinase K treatment in order to digest any substrate left on the outside of the vesicles. Protease-protected proOmpA that has been successfully translocated inside the lumen of the vesicles is detected by Western blotting.

Figure 12 In vitro translocation assay showing the effect of the proton-motive force (PMF) on secretion activity of SecYEG and the HTL. The complexes were coreconstituted together with bacteriorhodopsin (BR) in order to generate the PMF. BR-only vesicles were used as a negative control. Translocation reactions were incubated either in the presence (+Light) or in the absence (Light) of light, or in the presence of an uncoupling ionophore in the presence of light (+ CCCP). The graph shows percentage of proOmpA translocated inside the lumen of vesicles by SecYEG or the HTL. Quantification was performed using ImageJ software, and values represent the mean of three independent experiments SEM. Modified from Schulze et al. (2014).

ACEMBLing Multisubunit Membrane Protein Complexes

41

absence of the PMF (Light) was reduced compared to that of SecYEG, consistent with observations made during ATP-driven secretion with non-BR HTL vesicles. When BR-only vesicles were used in the PMFstimulated transport assay, no proOmpA was translocated, as expected. Taken together, the SecA- and PMF-stimulated translocation data presented above suggest that the HTL is surprisingly less capable of protein secretion compared to SecYEG. Nevertheless, it is more responsive to the presence of the PMF, as previously reported (Schulze et al., 2014). 4.3.3 Membrane protein insertion activity of the HTL The HTL complex contains both SecYEG and YidC as its components, which suggests that it might be active not only in protein secretion but also in membrane protein integration. For the purpose of comparing insertion efficiencies of the HTL and SecYEG, a model nascent membrane protein CyoA was chosen. CyoA is a subunit II of the cytochrome bo3 quinol oxidase complex, containing three transmembrane helices and a large periplasmic domain on its C-terminus. As reported by various studies, CyoA is believed to require both SecYEG and YidC for its efficient incorporation into the membrane (Celebi, Yi, Facey, Kuhn, & Dalbey, 2006; du Plessis, Nouwen, & Driessen, 2006). This makes it a good model substrate for studying activity of the HTL, which is composed of both of these proteins. In order to determine the capability of the HTL to insert membrane proteins, in vitro transcription followed by a coupled in vitro translation/insertion assay was performed (Fig. 13). cyoA mRNA generated in the process of in vitro transcription was first translated using an E. coli S30 cell extract in the presence of scSRP (single-chain SRP), 35S-labeled methionine and PLs containing either the SecYEG or the HTL complex and subsequently co-translationally incorporated into those vesicles. Following vesicle floatation, resistance to extraction by urea determined the amount of successfully inserted radiolabeled protein, detected by phosphorimaging (Fig. 14). The data were normalized to insertion activity of SecYEG, clearly showing that CyoA was more efficiently inserted into proteolipososmes containing the HTL, compared to SecYEG alone, as previously reported (Schulze et al., 2014). In the presence of empty vesicles, only a negligible amount of protein was detected, indicating the level of background spontaneous insertion of CyoA.

42

Joanna Komar et al.

Figure 13 In vitro translation/membrane protein insertion coupled assay setup. mRNA coding for a membrane protein of interest is translated using E. coli S30 cell extract in the presence of the complex of signal recognition particle (SRP) and its receptor FtsY (single-chain SRP—scSRP), 35S-methionine and proteoliposomes containing reconstituted translocation complexes, and subsequently co-translationally inserted into the vesicles. Following a 90-min incubation, vesicles are purified by sucrose density gradient centrifugation and treated with 5 M urea. Successfully inserted substrates are detected by phosphorimaging.

Figure 14 Coupled in vitro translation/insertion assay comparing membrane protein insertion activity of SecYEG and the HTL. (A) A diagram showing topology of the model membrane protein substrate CyoA. The arrow indicates the signal peptidase (SP) cleavage site. (B) In vitro translation/insertion assay of CyoA. cyoA mRNA was synthesized in vitro and subsequently translated using an E. coli S-30 cell extract in the presence of single-chain signal recognition particle (scSRP), 35S-labeled methionine and empty vesicles or proteoliposomes containing either SecYEG or the HTL. Following vesicle floatation and urea treatment, reactions were analyzed by SDS-PAGE and successfully inserted radiolabeled protein was detected by phosphorimaging. Quantification was performed using ImageQuant software, and values represent the mean of four independent experiments SEM. Modified from Schulze et al. (2014).

ACEMBLing Multisubunit Membrane Protein Complexes

43

5. DISCUSSION AND CONCLUSIONS A stable HTL complex, which comprises the core translocon SecYEG, the membrane protein insertase YidC and the poorly understood additional components SecDF-YajC, has been successfully expressed and purified (Schulze et al., 2014). Until recently, this has presented a virtually insurmountable challenge. Being in a position to overexpress and purify an intact complex with the ACEMBL technology and to reconstitute it, enables a thorough analysis of its activity in both post- and co-translational translocation. Results summarized here demonstrate that the HTL is capable of both of these processes. As expected due to the presence of SecYEG in the complex, the HTL is active in ATP-stimulated protein secretion (Fig. 10); however, it appears less efficient at this process compared to the core SecYEG translocon, despite having a higher affinity for SecA and a translocating substrate (Schulze et al., 2014). Nevertheless, the HTL-mediated transport is more dependent on the PMF (Fig. 12). This might result from the HTL compensating for the lower conversion efficiency of ATP to translocation, as both the HTL and SecYEG generate the same level of SecA ATPase activity corresponding to transport, however with a lower yield in proOmpA translocation for the HTL (Schulze et al., 2014). The higher stimulation of the HTL by the PMF might be mediated by SecD and SecF, homologues of PMF-driven transport proteins, like, e.g., AcrB drug transporter (Tsukazaki et al., 2011). Their role in preprotein translocation mediated by regulating membrane cycling of SecA ATPase (Duong & Wickner, 1997b; Economou, Pogliano, Beckwith, Oliver, & Wickner, 1995) might also play a role in this process. It is likely that the HTL utilizes the PMF through both SecYEG and SecDF, and that the high stimulation of the HTL-mediated proOmpA translocation is a cumulative effect. However, the mechanism of PMF stimulation in protein translocation is yet to be determined. The HTL complex is also capable of membrane protein insertion, being more efficient in this process compared to SecYEG alone for the tested substrate (Fig. 14B). This suggests that the nascent polypeptide chain emerging from a ribosome likely contacts both SecYEG and YidC in the HTL. This is supported by ribosome binding studies presented in Schulze et al., where all individual complexes: HTL, SecYEG, YidC, and SecDF-YajC-YidC, apart from SecDF alone, associated preferentially with translating ribosomes presenting the nascent TMS of FtsQ over nontranslating ribosomes.

44

Joanna Komar et al.

Additionally, fluorescence studies showed higher binding affinity of the HTL for the ribosomes compared to SecYEG (Schulze et al., 2014), which supports the higher efficiency of the HTL in co-translational membrane protein insertion. The higher level of CyoA insertion in the presence of the HTL is also supported by previous studies showing dependency of this substrate on both SecYEG and YidC for its efficient insertion (Celebi, Dalbey, & Yuan, 2008; du Plessis et al., 2006). CyoA has also been reported to strongly require YidC for its proper membrane integration (van der Laan et al., 2003), which could explain its low insertion levels with SecYEG alone. The presence of YidC in the HTL complex, which accounts for the increased efficiency of membrane protein insertion, might be essential for protein assembly and membrane biogenesis in vivo. However, the fact that the HTL also possesses the ability to export proteins across the membrane might be required for translocation of large periplasmic domains of membrane protein substrates, like that of CyoA (Fig. 14A). Our results suggest that the HTL might be preferential for membrane protein insertion, whereas SecYEG alone might be mainly utilized in protein secretion. In vivo however, the requirement for each of those complexes is probably determined by the nature of a translocating substrate, resulting in the observed versatility of the Sec machinery. As the individual components of the HTL can act alone in protein secretion and membrane protein insertion, assembly of the HTL complex is most likely a very dynamic process in vivo. This is supported by the fact that the evaluated number of copies of different translocons are very diverse. SecYE is believed to exist in a cell in about 300–400 copies, SecDF in only about 30 copies (Pogliano & Beckwith, 1994), and YidC in a substantially higher number of 2500 copies per cell (Urbanus et al., 2002). This suggests that for every HTL complex, consisting of single copies of SecYEG and SecDF-YajC-YidC, there might be up to five copies of the SecYEG dimer. The much higher estimated number of YidC copies could be explained by its proposed independent roles as an insertase (Serek et al., 2004; van der Laan, Bechtluft, Kol, Nouwen, & Driessen, 2004), a chaperone (Beck et al., 2001; Nagamori, Smirnova, & Kaback, 2004), and an assembly site (Pop et al., 2009), as well as its involvement in degradation of misfolded proteins in association with a partner protease FtsH (van Bloois et al., 2008). Existence of various translocons, consisting of either one subunit of SecYEG and SecDF-YajC-YidC, two SecYEG monomers or YidC only, might be determined by different stages of cell growth or by various

ACEMBLing Multisubunit Membrane Protein Complexes

45

environmental conditions. An analogous phenomenon also exists in eukaryotic organelles. The Sec61 translocon, found in the endoplasmic reticular membrane, forms complexes with its accessory proteins, like the Sec62/63 complex (Meyer et al., 2000; Reithinger, Kim, & Kim, 2013). Moreover, based on EM studies, it has been proposed that the Sec61 complex forms oligomers upon interaction with the Sec62/63 complex or ribosomes (Hanein et al., 1996). Similarly, the mitochondrial import complexes are likely to associate with additional factors, e.g., Oxa1, a homologue of the bacterial YidC insertase. Successful purification of the HTL complex is a breakthrough in the protein transport field. Not only has it enabled analysis of the HTL’s activity in protein secretion and membrane protein insertion, but it has also provided means for studying subunit interactions within the complex, as well as for its structural analysis. By using a cross-linking approach, higher MW complexes were detected in the HTL sample, but not in the SecYEG sample (Fig. 8). These cross-links correspond to SecG–SecD and SecG–YidC interactions, as verified by mass spectrometry (Schulze et al., 2014). We have described here the production, purification, and functional analysis of the SecYEG-SecDF-YajC-YidC HTL supercomplex. Our work was made possible by our development of the new and powerful recombinant expression technology, ACEMBL, which harvests recombinationbased DNA assembly approaches for the rapid and flexible construction of multigene expression constructs that allow production of functional multisubunit membrane protein complexes. We have used ACEMBL to efficiently produce a prokaryotic membrane protein complex in E. coli as an expression host. Many membrane protein complexes will require eukaryotic expression systems for their efficient production for molecular level functional studies. The ACEMBL technology concept is not limited to expression in E. coli. The same approach can likewise be used in eukaryotic systems, and we have developed and made available the corresponding DNA reagents also for mammalian and baculovirus/insect cell systems (Bieniossek et al., 2012; Trowitzsch et al., 2010; Vijayachandran et al., 2011). The baculovirus/insect cell system has been exceptionally useful to produce G-protein-coupled receptors (Tate & Schertler, 2009). We anticipate that concepts such as those underlying ACEMBL will be invaluable to unlock the structure and function of multiprotein assemblies that constitute GPCR-dependent pathways and signaling cascades in the cell. Likewise, we imagine that technologies such as ACEMBL may also accelerate the rational design and construction of synthetic signaling pathways and complex

46

Joanna Komar et al.

assemblies that recapitulate multicomponent cellular mechanisms in in vitro and in vivo systems, providing manifold new opportunities for cell-based assay development and even the rational design of advanced molecular diagnostic and drug discovery tools. The important advancement in the field of protein translocation research, which we describe here, featuring the SecYEG-SecDF-YajCYidC HTL supercomplex as a prominent example, will contribute to a more refined structure of protein secretion and membrane protein insertion machinery in the future. For instance, the HTL can now be analyzed with a variety of trapped substrates. Our work also sets the stage for elucidating the molecular architecture of the HTL supercomplex by high-resolution structural analysis, by electron cryomicroscopy, and X-ray crystallography, ultimately providing an atomic resolution understanding of the plethora of subunit interactions within the HTL complex and their functional roles in catalyzing protein transport at the cell membrane.

ACKNOWLEDGMENTS The authors thank Sir John Walker for the E. coli C43 expression strain, Dr. John Bason for help with reconstitution of bacteriorhodopsin, Dr. Ryan Schulze for important contributions to the early stages of the project and all members of the Schaffitzel, Collinson and Berger laboratories for helpful discussions. J. K. was supported by a doctoral training grant from the BBSRC. C. S. is supported by a European Research Council ERC Starting Grant Award. I. C. acknowledges support by the BBSRC (Project Grants BB/M003604/1 and BB/I008675/1) and is recipient of a Wellcome Trust Senior Investigator Award. I. B. acknowledges support from the European Commission Framework Programme 7 ComplexINC project (contract no. 279039).

REFERENCES Akiyama, Y., & Ito, K. (1990). Secy protein, a membrane-embedded secretion factor of E. coli, is cleaved by the ompt protease invitro. Biochemical and Biophysical Research Communications, 167(2), 711–715. Arkowitz, R. A., & Wickner, W. (1994). SecD and SecF are required for the proton electrochemical gradient stimulation of preprotein translocation. EMBO Journal, 13, 954–963. Barford, D., Takagi, Y., Schultz, P., & Berger, I. (2013). Baculovirus expression: Tackling the complexity challenge. Current Opinion in Structural Biology, 23(3), 357–364. Beck, K., Eisner, G., Trescher, D., Dalbey, R. E., Brunner, J., & Muller, M. (2001). YidC, an assembly site for polytopic Escherichia coli membrane proteins located in immediate proximity to the SecYE translocon and lipids. EMBO Reports, 2(8), 709–714. Bessonneau, P., Besson, V., Collinson, I., & Duong, F. (2002). The SecYEG preprotein translocation channel is a conformationally dynamic and dimeric structure. EMBO Journal, 21(5), 995–1003. Bieniossek, C., Imasaki, T., Takagi, Y., & Berger, I. (2012). MultiBac: Expanding the research toolbox for multiprotein complexes. Trends in Biochemical Sciences, 37(2), 49–57.

ACEMBLing Multisubunit Membrane Protein Complexes

47

Bieniossek, C., Nie, Y., Frey, D., Olieric, N., Schaffitzel, C., Collinson, I., et al. (2009). Automated unrestricted multigene recombineering for multiprotein complex production. Nature Methods, 6(6), 447–450. Boy, D., & Koch, H. G. (2009). Visualization of distinct entities of the SecYEG translocon during translocation and integration of bacterial proteins. Molecular Biology of the Cell, 20(6), 1804–1815. Brundage, L., Hendrick, J. P., Schiebel, E., Driessen, A. J., & Wickner, W. (1990). The purified E. coli integral membrane protein SecY/E is sufficient for reconstitution of SecA-dependent precursor protein translocation. Cell, 62(4), 649–657. Celebi, N., Dalbey, R. E., & Yuan, J. (2008). Mechanism and hydrophobic forces driving membrane protein insertion of subunit II of cytochrome bo 3 oxidase. Journal of Molecular Biology, 375(5), 1282–1292. Celebi, N., Yi, L., Facey, S. J., Kuhn, A., & Dalbey, R. E. (2006). Membrane biogenesis of subunit II of cytochrome bo oxidase: Contrasting requirements for insertion of N-terminal and C-terminal domains. Journal of Molecular Biology, 357(5), 1428–1436. Collinson, I., Breyton, C., Duong, F., Tziatzios, C., Schubert, D., Or, E., et al. (2001). Projection structure and oligomeric properties of a bacterial core protein translocase. EMBO Journal, 20(10), 2462–2471. Deville, K., Gold, V. A., Robson, A., Whitehouse, S., Sessions, R. B., Baldwin, S. A., et al. (2011). The oligomeric state and arrangement of the active bacterial translocon. The Journal of Biological Chemistry, 286(6), 4659–4669. Driessen, A. J. M. (1992). Precursor protein translocation by the Escherichia-coli translocase is directed by the protonmotive force. EMBO Journal, 11(3), 847–853. du Plessis, D. J., Nouwen, N., & Driessen, A. J. (2006). Subunit a of cytochrome o oxidase requires both YidC and SecYEG for membrane insertion. Journal of Biological Chemistry, 281(18), 12248–12252. Duong, F., & Wickner, W. (1997a). Distinct catalytic roles of the SecYE, SecG and SecDFyajC subunits of preprotein translocase holoenzyme. EMBO Journal, 16(10), 2756–2768. Duong, F., & Wickner, W. (1997b). The SecDFyajC domain of preprotein translocase controls preprotein movement by regulating SecA membrane cycling. EMBO Journal, 16(16), 4871–4879. Economou, A., Pogliano, J., Beckwith, J., Oliver, D., & Wickner, W. (1995). SecA membrane cycling at SecYEG is driven by distinct ATP binding and hydrolysis events and is regulated by SecD and SecF. Cell, 83(7), 1171–1181. Fancy, D. A., & Kodadek, T. (1999). Chemistry for the analysis of protein-protein interactions: Rapid and efficient cross-linking triggered by long wavelength light. Proceedings of the National Academy of Sciences of the United States of America, 96(11), 6020–6024. Fitzgerald, D. J., Berger, P., Schaffitzel, C., Yamada, K., Richmond, T. J., & Berger, I. (2006). Protein complex expression by using multigene baculoviral vectors. Nature Methods, 3(12), 1021–1032. Geller, B. L., Movva, N. R., & Wickner, W. (1986). Both ATP and the electrochemical potential are required for optimal assembly of pro-OmpA into Escherichia coli inner membrane vesicles. Proceedings of the National Academy of Sciences of the United States of America, 83(12), 4219–4222. Gold, V. A., Robson, A., Bao, H., Romantsov, T., Duong, F., & Collinson, I. (2010). The action of cardiolipin on the bacterial translocon. Proceedings of the National Academy of Sciences of the United States of America, 107(22), 10044–10049. Haffke, M., Viola, C., Nie, Y., & Berger, I. (2013). Tandem recombineering by SLIC cloning and Cre-LoxP fusion to generate multigene expression constructs for protein complex research. Methods in Molecular Biology, 1073, 131–140.

48

Joanna Komar et al.

Hanein, D., Matlack, K., Jungnickel, B., Plath, K., Kalies, K., Miller, K., et al. (1996). Oligomeric rings of the Sec61p complex induced by ligands required for protein translocation. Cell, 87(4), 721–732. Kato, Y., Nishiyama, K., & Tokuda, H. (2003). Depletion of SecDF-YajC causes a decrease in the level of SecG: Implication for their functional interaction. FEBS Letters, 550(1–3), 114–118. Kumazaki, K., Chiba, S., Takemoto, M., Furukawa, A., Nishiyama, K., Sugano, Y., et al. (2014). Structural basis of Sec-independent membrane protein insertion by YidC. Nature, 509(7501), 516–520. Li, M. Z., & Elledge, S. J. (2007). Harnessing homologous recombination in vitro to generate recombinant DNA via SLIC. Nature Methods, 4(3), 251–256. Luirink, J., von Heijne, G., Houben, E., & de Gier, J. W. (2005). Biogenesis of inner membrane proteins in Escherichia coli. Annual Review of Microbiology, 59, 329–355. Meyer, H. A., Grau, H., Kraft, R., Kostka, S., Prehn, S., Kalies, K. U., et al. (2000). Mammalian Sec61 is associated with Sec62 and Sec63. The Journal of Biological Chemistry, 275(19), 14550–14557. Miroux, B., & Walker, J. E. (1996). Over-production of proteins in Escherichia coli: Mutant hosts that allow synthesis of some membrane proteins and globular proteins at high levels. Journal of Molecular Biology, 260(3), 289–298. Nagamori, S., Smirnova, I. N., & Kaback, H. R. (2004). Role of YidC in folding of polytopic membrane proteins. Journal of Cell Biology, 165(1), 53–62. Pogliano, K. J., & Beckwith, J. (1994). Genetic and molecular characterization of the Escherichia coli secD operon and its products. Journal of Bacteriology, 176(3), 804–814. Pop, O. I., Soprova, Z., Koningstein, G., Scheffers, D. J., van Ulsen, P., Wickstrom, D., et al. (2009). YidC is required for the assembly of the MscL homopentameric pore. The FEBS Journal, 276(17), 4891–4899. Reithinger, J. H., Kim, J. E., & Kim, H. (2013). Sec62 protein mediates membrane insertion and orientation of moderately hydrophobic signal anchor proteins in the endoplasmic reticulum (ER). The Journal of Biological Chemistry, 288(25), 18058–18067. Robson, A., Booth, A. E., Gold, V. A., Clarke, A. R., & Collinson, I. (2007). A large conformational change couples the ATP binding site of SecA to the SecY protein channel. Journal of Molecular Biology, 374(4), 965–976. Sachelaru, I., Petriman, N. A., Kudva, R., Kuhn, P., Welte, T., Knapp, B., et al. (2013). YidC occupies the lateral gate of the SecYEG translocon and is sequentially displaced by a nascent membrane protein. The Journal of Biological Chemistry, 288(23), 16295–16307. Schiebel, E., Driessen, A. J., Hartl, F. U., & Wickner, W. (1991). Delta mu H + and ATP function at different steps of the catalytic cycle of preprotein translocase. Cell, 64(5), 927–939. Schulze, R. J., Komar, J., Botte, M., Allen, W. J., Whitehouse, S., Gold, V. A., et al. (2014). Membrane protein insertion and proton-motive-force-dependent secretion through the bacterial holo-translocon SecYEG-SecDF-YajC-YidC. Proceedings of the National Academy of Sciences of the United States of America, 111(13), 4844–4849. Scotti, P. A., Urbanus, M. L., Brunner, J., de Gier, J. W., von Heijne, G., van der Does, C., et al. (2000). YidC, the Escherichia coli homologue of mitochondrial Oxa1p, is a component of the Sec translocase. EMBO Journal, 19(4), 542–549. Serek, J., Bauer-Manz, G., Struhalla, G., van den Berg, L., Kiefer, D., Dalbey, R., et al. (2004). Escherichia coli YidC is a membrane insertase for Sec-independent proteins. EMBO Journal, 23(2), 294–301. Shiozuka, K., Tani, K., Mizushima, S., & Tokuda, H. (1990). The proton motive force lowers the level of ATP required for the invitro translocation of a secretory protein in Escherichia-coli. Journal of Biological Chemistry, 265(31), 18843–18847.

ACEMBLing Multisubunit Membrane Protein Complexes

49

Tate, C. G., & Schertler, G. F. (2009). Engineering G protein-coupled receptors to facilitate their structure determination. Current Opinion in Structural Biology, 19(4), 386–395. Trowitzsch, S., Bieniossek, C., Nie, Y., Garzoni, F., & Berger, I. (2010). New baculovirus expression tools for recombinant protein complex production. Journal of Structural Biology, 172(1), 45–54. Tsukazaki, T., Mori, H., Echizen, Y., Ishitani, R., Fukai, S., Tanaka, T., et al. (2011). Structure and function of a membrane component SecDF that enhances protein export. Nature, 474(7350), 235–238. Urbanus, M. L., Froderberg, L., Drew, D., Bjork, P., de Gier, J. W., Brunner, J., et al. (2002). Targeting, insertion, and localization of Escherichia coli YidC. Journal of Biological Chemistry, 277(15), 12718–12723. Urbanus, M. L., Scotti, P. A., Froderberg, L., Saaf, A., de Gier, J. W., Brunner, J., et al. (2001). Sec-dependent membrane protein insertion: Sequential interaction of nascent FtsQ with SecY and YidC. EMBO Reports, 2(6), 524–529. van Bloois, E., Dekker, H. L., Froderberg, L., Houben, E. N., Urbanus, M. L., de Koster, C. G., et al. (2008). Detection of cross-links between FtsH, YidC, HflK/C suggests a linked role for these proteins in quality control upon insertion of bacterial inner membrane proteins. FEBS Letters, 582(10), 1419–1424. van den Berg, B., Clemons, W. M., Jr., Collinson, I., Modis, Y., Hartmann, E., Harrison, S. C., et al. (2004). X-ray structure of a protein-conducting channel. Nature, 427(6969), 36–44. van der Laan, M., Bechtluft, P., Kol, S., Nouwen, N., & Driessen, A. J. (2004). F1F0 ATP synthase subunit c is a substrate of the novel YidC pathway for membrane protein biogenesis. Journal of Cell Biology, 165(2), 213–222. van der Laan, M., Urbanus, M. L., Ten Hagen-Jongman, C. M., Nouwen, N., Oudega, B., Harms, N., et al. (2003). A conserved function of YidC in the biogenesis of respiratory chain complexes. Proceedings of the National Academy of Sciences of the United States of America, 100(10), 5801–5806. Vijayachandran, L. S., Viola, C., Garzoni, F., Trowitzsch, S., Bieniossek, C., Chaillet, M., et al. (2011). Robots, pipelines, polyproteins: Enabling multiprotein expression in prokaryotic and eukaryotic cells. Journal of Structural Biology, 175(2), 198–208. Xie, K., Kiefer, D., Nagler, G., Dalbey, R. E., & Kuhn, A. (2006). Different regions of the nonconserved large periplasmic domain of Escherichia coli YidC are involved in the SecF interaction and membrane insertase activity. Biochemistry, 45(44), 13401–13408.