Nanodiscs and SILAC-Based Mass Spectrometry to Identify a

Dec 1, 2011 - Department of Biochemistry and Molecular Biology, University of British Columbia, Vancouver, BC, Canada. ‡. Centre for High-Throughput...
0 downloads 0 Views 2MB Size
Technical Note pubs.acs.org/jpr

Nanodiscs and SILAC-Based Mass Spectrometry to Identify a Membrane Protein Interactome Xiao X. Zhang,† Catherine S. Chan,† Huan Bao,† Yuan Fang,†,‡ Leonard J. Foster,†,‡ and Franck Duong*,† †

Department of Biochemistry and Molecular Biology, University of British Columbia, Vancouver, BC, Canada Centre for High-Throughput Biology, University of British Columbia, Vancouver, BC, Canada



S Supporting Information *

ABSTRACT: Integral membrane proteins are challenging to work with biochemically given their insoluble nature; the nanodisc circumvents the difficulty by stabilizing them in small patches of lipid bilayer. Here, we show that nanodiscs combined with SILACbased quantitative proteomics can be used to identify the soluble interacting partners of virtually any membrane protein. As a proof of principle, we applied the method to the bacterial SecYEG protein-conducting channel, the maltose transporter MalFGK2 and the membrane integrase YidC. In contrast to the detergent micelles, which tend to destabilize interactions, the nanodisc was able to capture out of a complex whole cell extract the proteins SecA, Syd, and MalE with a high degree of confidence and specificity. The method was sensitive enough to isolate these interactors as a function of the lipid composition in the disc and the culture conditions. In agreement with a previous photo-cross linking analysis, YidC did not show any high-affinity interactions with cytosolic or periplasmic proteins. These three examples illustrate the utility of nanoscale lipid bilayers to identify the soluble peripheral partners of proteins intergrated in the lipid bilayer. KEYWORDS: nanodiscs, SILAC, SecYEG, MalFGK2, YidC, membrane proteome, protein−lipid interactions



INTRODUCTION Membrane proteins constitute nearly 30% of the cellular proteome in most organisms.1 These proteins carry out important functions such as energy production, signal transduction, protein trafficking, molecular transport, and host− pathogen interactions. Despite their importance, the membraneinteracting proteome remains largely unexplored, partly because of its marked insolubility and tendency to aggregate when extracted from the hydrophobic protein−lipid bilayer environment. Proteomic-based methods relying on membrane vesicles (liposomes) and detergent micelles are particularly challenging.2 Membrane vesicles are heterogeneous in size and shape and often create hydrophobic crevasses that promote nonspecific associations. Detergents are useful to extract membrane proteins from the lipid bilayer environment, yet even the mildest ones often destabilize protein complexes during affinity pull-down, gel chromatography, and blue-native gel electrophoresis analysis.3,4 The identification of these membrane protein complexes is even more difficult when the interactions are characterized by low affinities or depend on the presence of specific lipids. It is thus not surprising that the proteomic efforts applied to the membrane proteome have generated limited set of data with limited degree of specificity.2,5−7 Furthermore, the proteome that is peripherally bound to the membrane, either via lipid or protein interactions, cannot be predicted using bioinformatic approaches, and its identification depends on experimental analysis carried on an individual basis. © 2011 American Chemical Society

In this report, we present a method that combines nanodiscs and SILAC labeling to identify the soluble interacting partners of membrane embedded proteins. Nanodiscs are small, water-soluble and homogeneous particles (∼10 nm-wide) that stabilize membrane proteins into a defined lipid environment.8 The technology depends on an engineered ∼200 residue amphipathic membrane scaffold protein (MSP), whose hydrophobic faces circumscribe the edges of a small circular lipid bilayer (Figure 1). Hence, analytical procedures such as affinity pull-down can be done in the absence of detergent. Stable isotope labeling by amino acids in cell culture (SILAC) is an in vivo protein-labeling technique developed for quantitative proteomics. Typically, a cell culture is grown with “lightunlabeled” or “heavy-isotope labeled” amino acids, resulting in two identical proteomes that differ only slightly in mass.9 Following incubation with nanodiscs containing the target membrane protein, the potential interactors are isolated by affinity pull-down and identified by tandem mass-spectrometry. The light-to-heavy ratio measured for every isolated peptide reflects the abundance of the protein in the sample and therefore the specificity of the interaction. As often is the case, the bacterial envelope proteome is an excellent test bed for developing novel methodology. Dozens of membrane proteins with known and unknown functions (“orphans”) are prime candidates for incorporation into discs Received: August 30, 2011 Published: December 1, 2011 1454

dx.doi.org/10.1021/pr200846y | J. Proteome Res. 2012, 11, 1454−1459

Journal of Proteome Research

Technical Note

Sequencing grade trypsin was purchased from Roche (Laval, QC, Canada). Preparation of SILAC Soluble Prey Fractions

E. coli strain JW2806 [Δ(araD-araB)567 ΔlacZ4787(::rrnB-3) lambda− ΔlysA763::kan rph-1 Δ(rhaD-rhaB)568 hsdR514]10 containing the lysA knockout was grown in LB media overnight at 37 °C. Following two washes in M9 minimal media, the culture was diluted 1:100 (relative to the original culture) into M9 minimal media supplemented with 2 mM magnesium sulfate, 0.4% w/v glucose or 0.2% w/v maltose, 10 μM CaCl2, 4 μg/mL of vitamin B1, and 0.2 mg/mL of all amino acids except the labeled cultures, which contained 0.3 mg/mL of 13C615N2lysine and 0.174 mg/mL of 13C615N4-arginine. Cultures were grown for approximately 6 h at 37 °C, which is equivalent to six generations. Cells were solubilized in TSG buffer (50 mM TrisHCl pH 7.9, 100 mM NaCl, 10% w/v glycerol) and lysed by three passes in a French pressure cell. The debris was removed by low-speed centrifugation (3000 rcf, 10 min), and the soluble fraction was isolated by ultracentrifugation (126 000 rcf, 45 min). The unlabeled and heavy-labeled protein fractions were stored in −70 °C at a protein concentration of ∼20−25 mg/mL (determined by a Bradford assay) prior to use in pull-down experiments. Protein and Nanodisc Preparation

The His6-tagged SecYEG complex was overexpressed, solubilized, and purified as described11 through immobilized metal affinity chromatography, followed by anion exchange chromatography. N-terminal His6-tagged YidC was overexpressed from the plasmid pBAD22YidC in E. coli BL21(DE3) cells, and the membrane was solubilized with n-dodecyl β-D-maltoside and purified under similar conditions as SecYEG. Further purification was achieved by gel filtration chromatography using a Tricorn Superdex 200 HR 10/300 column. MalFGK2 complex was overexpressed from the plasmid pBAD22-FGK bearing a C-terminal His6-tag on MalK in E. coli BL21(DE3) cells. Membranes were solubilized under similar conditions as SecYEG except at 5 mg/mL in buffer containing 20% instead of 10% glycerol and purified by immobilized metal affinity chromatography and gel filtration chromatography as above. Nanodiscs were reconstituted as described in Dalal and Duong.11 The SecEYG:MSP1:lipid ratio employed was 1:4:40. The reconstitution of YidC and MalFGK was performed without added lipids at a protein:MSP1 ratio of 1:4 and 1:5, respectively. The prepared discs were purified by gel filtration chromatography.11

Figure 1. Strategy employed to identify the interacting partners of a membrane protein using nanodisc and SILAC. Nanodiscs (right) containing a membrane protein of interest, or the His-tagged membrane scaffold protein alone as a control (left), were used as baits. Crude protein extract from cells grown in normal “light” or stable isotope “heavy” analogue media were used as prey. After affinity pull-down using Ni-NTA sepharose beads, the coeluted proteins were analyzed by SDS-PAGE and their identity revealed by LC−MS/MS. For each peptide identified, the peak intensity of the heavy form is compared to the light form to determine whether a protein is a specific interactor or nonspecific contaminant.

and for subsequent proteomic analysis using the principle described above. The results presented here employ the SecYEG channel, the maltose transporter MalFGK2, and the insertase YidC to demonstrate the utility of nanoscale lipid bilayers for membrane proteomic analysis.



Pull-Down Experiments

The indicated nanodiscs (10 μg each), or the scaffold protein MSP1 as control, were immobilized onto Ni2+-NTA sepharose beads (20 μL suspension) in 0.5 mL of TSG buffer for 5 min at room temperature with gentle shaking. After washing with 0.5 mL of TSG buffer to remove excess unbound protein, the beads were incubated with the unlabeled or labeled soluble protein fractions (1 mg each) in 0.5 mL of TSG buffer. Beads from both samples were then combined together and washed three times in 1 mL of TSG buffer containing 50 mM imidazole, followed by elution with 0.1 mL of TSG buffer containing 600 mM imidazole. The pull-down efficiency was assessed by 12% SDSPAGE and silver staining prior to digestion and analysis of the sample by mass spectrometry.

EXPERIMENTAL SECTION

Materials

Amino acid isotopologues were purchased from Cambridge Isotope Laboratories, (Andover, MA) and all others from Sigma-Aldrich (St. Louis, MO). Dioleoyl-sn-glycero-3-phospho(1′-rac-glycerol) (DOPG) and Escherichia coli total lipid extract were purchased from Avanti Polar Lipids (Alabaster, AL). Ni2+NTA chelating Sepharose and Superdex 200 bead suspensions were purchased from GE Life Sciences (Uppsala, Sweden). 1455

dx.doi.org/10.1021/pr200846y | J. Proteome Res. 2012, 11, 1454−1459

Journal of Proteome Research

Technical Note

Table 1. Top Ten Peptide Ratio Identified Using Nanodiscs Containing SecYEG, MalFGK2, and YidCa accession no.b

name

SecYEG(Ec) P0A8U0 Syd P0A6T5 FolE P0A825 GlyA P06987 HisB P09373 PflB P0ACK8 FucR P64588 YqjL P0A8P6 XerC P0ACJ8 Crp P0ACE7 HinT SecYEG(PG) P10408 SecA P0A8P6 XerC P08622 DnaJ P0A8U0 Syd P0ACE7 HinT P04949 FliC P0A722 LpxA P0A9A9 Fur P0ACJ8 Crp P0A951 SpeG MalFGK(Glu) P0AEX9 MalE P0AFY8 SeqA P0A7V0 RpsB P62768 YaeH P0AFI5 PbpG

description/function

ratioc

SecY-interacting protein GTP cyclohydrolase I serine hydroxymethyltransferase histidine biosynthesis bifunctional protein formate acetyltransferase L-fucose operon activator uncharacterized protein chromosome segregation recombinase cAMP receptor protein purine nucleoside phosphoramidase

27 4.2 3.9 3.4 3.2 2.8 2.8 2.7 1.5 1.2

preprotein translocase ATPase chromosome segregation recombinase chaperone/heat shock protein SecY-interacting protein purine nucleoside phosphoramidase flagellin UDP N-acetylglucosamine acyltransferase ferric uptake regulator cAMP receptor protein spermidine acetyltransferase

35 9.0 6.3 4.7 3.7 3.2 3.2 3.0 2.7 2.6

periplasmic maltose binding protein DNA replication regulation modulator of oriC 30S ribosomal protein S2 function unknown, UPF0325 family protein D-alanyl-D-alanine endopeptidase

accession no.b

name

MalFGK(Glu) P0ACI0 Rob P0ACJ8 Crp P0AFG0 NusG P42596 RlmG P0A8P8 XerD MalFGK(Mal) P0AEX9 MalE P62768 YaeH P0AEZ3 MinD P0A7N9 RpmG P0A7L8 RpmA P0ADZ4 RpsO P0A7M2 RpmB P0A7N4 RpmF P36548 AmiA P0A6X3 Hfq YidC P0A8P6 XerC P0ACK8 FucR P68066 GrcA P09373 PflB P0A825 GlyA P0ACE7 HinT P0A6T5 FolE P06987 HisB P0ACI0 Rob P0A7F3 PyrI

4.4 3.4 2.4 2.2 2.1

description/function right origin-binding protein cAMP receptor protein transcription antitermination protein 23S rRNA methyltransferase chromosome segregation recombinase

ratioc 1.8 1.8 1.5 1.5 1.5

periplasmic maltose binding protein function unknown, UPF0325 family protein septum site determination protein 50S ribosomal protein L33 50S ribosomal protein L27 30S ribosomal protein S15 50S ribosomal protein L28 50S ribosomal protein L32 N-acetylmuramyl-L-alanine amidase host-factor I protein

26 2.4 1.6 1.2 1.1 1.1 1.0 1.0 1.0 0.9

chromosome segregation recombinase operon activator autonomous glycyl radical cofactor formate acetyltransferase serine hydroxymethyltransferase purine nucleoside phosphoramidase GTP cyclohydrolase I histidine biosynthesis bifunctional protein right origin-binding protein aspartate carbamoyltransferase

5.2 5.1 3.9 3.3 3.2 3.2 3.0 2.7 1.9 1.8

L-fucose

a

The SecYEG complex was reconstituted with E. coli total lipids (Ec) or with phosphatidylglycerol (PG). The MalFGK2 complex was incubated with cytosolic extract from culture grown with glucose (Glu) or maltose (Mal). The complete list is provided in the Supporting Information. bSwiss-Prot accession numbers. cAverage heavy-to-light peptide ratios for each protein from one to three replicates.

Mass Spectrometry

one repetition with scores greater than 25 were considered identified, on the basis of an estimation of the false discovery rate of such a filter yielding 1 were identified (Figure 3A, Table 1 and Supporting Information Table S1 to compare reproducibility across three different experiments). Inspection of the results revealed the natural variability of the SILAC ratio around or below ∼3.0. We therefore did not consider these lower ratios, especially if the captured proteins were not identified across independent experiments. With these specificity criteria, the protein Syd was found across three replicates and with a SILAC ratio >25, which was well above the ratio obtained for any other proteins captured in the same experiment (Figure 3A). The interaction was therefore highly specific and remarkable because Syd is not a particularly abundant protein in E. coli.16 In addition, previous biochemical analysis in detergent solution showed that Syd does not form a stable complex with SecY but instead disassembles SecYEG into individual subunits.17 Here, the incorporation of the SecYEG complex in nanodiscs allowed for the efficient and specific capture of Syd. The SecA ATPase was present in the list of apparent interactors but with a low peptide ratio (∼1.5), which was surprising considering that SecA displays nanomolar affinity for the SecY channel embedded in the membrane.18 We thus tested the possibility that membrane lipids contribute to the association of SecA with SecYEG, especially since acidic lipids are known to support the SecA ATPase activity.19 When the SecYEG complex was reconstituted in nanodisc in the presence of phosphatidylglycerol (PG), SecA was captured with a SILAC ratio >35, which was well above any other SILAC ratio in the same experiment (Figure 3B, Table 1, and Supporting Information Table S2). Syd was also identified but with a lower SILAC ratio (Figure 3B, position 4), probably because SecA and Syd compete for the same binding site on SecY, as shown previously.17 It is remarkable that neither FtsY (∼500 copies per cell) nor the ribosomal proteins L23 and L35 (>20 000 copies per cell) were captured in these experiments. The association of these proteins with the SecY channel is weak compared to SecA.20,21 In addition, the interaction depends on the formation of a ribosome-nascent chain complex during cotranslational translocation.20,21 Two other proteins, XerC and the chaperone DnaJ, were identified with SILAC ratio ∼5. The capture of XerC was mostly nonspecific, given the variability of the SILAC ratio and the capture of this protein in other pulldown experiments (Supporting Information Tables S2 and S5).

Application to the Maltose Transporter MalFGK2

We employed the ABC transporter MalFGK2 to attempt the capture of periplasmic proteins, in this case the maltose-binding protein MalE. A crude cell extract from E. coli grown with maltose was incubated with MalFGK2 reconstituted in nanodiscs (Figure 2). As above, the potential interacting partners were isolated by affinity pull-down and identified by LC−MS/MS. As expected, MalE was at the top of the list with a peptide ratio >26, well above the SILAC ratio obtained for other proteins in these experiments (Figure 3C, Table 1, and Supporting Information Table S3). This result was remarkable because the association of MalE with MalFGK2 was detected in membrane and detergent solution only under specific conditions (e.g., in the presence of vanadate or with nonhydrolyzable ATP analogues).22 Here, the nanodisc captured MalE and out of complex crude cell extract. The second protein in the list of interactions was YaeH, which is a protein that displays affinity for membrane lipids.5 The peptide ratio for YaeH was near background levels (∼2.4; Table 1), suggesting nonspecific interaction with the maltose transporter. Finally, neither MalQ , MalP, or MalT, whose synthesis is increased in the presence of maltose,23 were captured in these experiments, indicating that the binding of MalE to the disc is not simply a result of protein overproduction. To confirm this, we tested the capture of MalE using cells grown on glucose, to repress expression of the mal regulon.24 In these conditions, the overall SILAC ratios were near background levels but MalE was still at the top of the list of captured proteins (Figure 3C). It has been proposed that MalK may interact with the transcription regulator MalT and the PEP-enzyme EIIA when cells are grown with glucose, but this was based on indirect evidence.25,26 Here, we were able to confirm that these interactions, if they exist, are weak or transient because these proteins were not detected in the pulldown assays. Application to the Membrane Insertase YidC

The insertase YidC (Oxa1p in mitochondria) is a membraneembedded chaperone essential for the biogenesis of inner membrane proteins. YidC consists of six transmembrane segments and a large ∼300 amino acid long periplasmic domain, termed P1.27 The crystal structure of P1 revealed a protein fold resembling a carbohydrate-binding protein and a cleft motif that could serve as a peptide-binding site.28,29 Earlier proteomic analysis using in vitro photo-cross linking approach captured a dozen different soluble proteins, but none of them with an evident relationship to YidC.30 Here, YidC was purified to homogeneity and reconstituted in nanodiscs (Figure 2). A few cytosolic and periplasmic proteins were captured but none 1457

dx.doi.org/10.1021/pr200846y | J. Proteome Res. 2012, 11, 1454−1459

Journal of Proteome Research

Technical Note

proteomic analyses usually rely on amphipathic detergent micelles or organic solvent that solubilize membrane lipids and maintain membrane proteins in solution.2,4,31−33 Many interactions occurring at a membrane interface are however very sensitive to surfactants, leading to the destabilization or disassembly of the complexes into separate subunits. This is the case for SecA, Syd, and MalE, which are readily dissociated from their transporter following membrane solubilization with the mildest possible detergents. Furthermore, even though detergents are employed at levels near their critical micellar concentration, these peripheral proteins do not reassociate spontaneously with their membrane protein partners.17,22,34 To avoid detergents, binding assays have been performed using the extramembranar portions of the membrane protein. This method is not always possible however and the possibility exists of losing the correct strucural fold essential for protein interactions. Cross-linking has been an alternative method that has sometimes proven successful either in vivo or in membrane vesicle analysis.35,36 The identification of the cross linked products is problematic however, especially when the cross link is inefficient or poorly selective. The nanodisc has the potential to circumvent many of these difficulties as it recreates a membrane-like environment, preserves the structure of the membrane proteins, and converts the target protein into a soluble particle amenable to biochemical analysis. In addition, the lipid composition in the disc can be modified during the reconstitution process, which was a critical parameter for the successful isolation of SecA as a bona fide SecYEG interactor. The nanodisc is therefore an excellent tool to identify interacting partners but also to test whether a specific lipidic environment is critical for protein binding. Although the method has great potential and can virtually be applied to any membrane protein incorporated into the disc, the success may still depend on other parameters such as growth or binding conditions. For example, in the case of MalFGK2, it was possible to capture MalE out of a cell extract grown with glucose, although the SILAC ratio and the degree of confidence were much higher when cells were grown with maltose, most likely as a result of MalE increase in the cellular extract. In contrast, very abundant ribosomal proteins were not captured with SecYEG, most likely because these interactions depends on specific conditions such as cotranslational protein translocation. The method may also be limited when an interactor is strongly or permanently bound to the lipid bilayer (e.g., with a lipid anchor) and therefore unlikely be released into the cytosolic cellular extract. On the other hand, some other proteins such as XerC and FucR may have the propensity to associate with the disc while not being true interactors. In this regard, the SILAC ratio across three replicates is a crucial indicator of the interaction specificity: the higher the ratio, the better the confidence. In conclusion, the method is a valuable addition to the membrane proteomics toolbox yet, as with many proteomicbased detection requires its complement of biochemical and biological techniques.

Figure 3. Average peptide ratio for the proteins identified by LC−MS/ MS analysis. The higher the ratio, the higher the confidence. The nanodiscs included in this analysis were the SecYEG translocon reconstituted with E. coli total lipids (Ec lipids) or phosphatidylglycerol (PG lipids), the MalFGK transporter, and the YidC insertase. Pull-down experiments with the MalFGK transporter were using protein extracts generated from cells grown in glucose or maltose, as indicated. The bars represent the average peptide ratio heavy-to-light for each identified protein. Proteins are numbered according to Table 1. The complete list of identified proteins is provided in Supplementary Tables 1 to 5.

with SILAC ratio >6 (Figure 3D and Supporting Information Table S5). The proteins XerC and FucR were at the top of the list but also captured in experiments using SecYEG and MalFGK2, suggesting a nonspecific interaction. We concluded that the insertase YidC does not have high-affinity interaction with cytosolic or periplasmic partners.





DISCUSSION Our results show that the nanodisc can be successfully employed in proteomic analysis to capture the peripheral interacting partners of virtually any target membrane protein. When combined with SILAC-labeling, the method is very sensitive and can detect true interactors out of a whole cell extract containing thousands of different protein species. Other

ASSOCIATED CONTENT

S Supporting Information *

Tables S1−S5, protein and peptide ratios obtained under different conditions. This material is available free of charge via the Internet at http://pubs.acs.org. 1458

dx.doi.org/10.1021/pr200846y | J. Proteome Res. 2012, 11, 1454−1459

Journal of Proteome Research



Technical Note

binding, and activity of Syd, a SecY-interacting protein. J. Biol. Chem. 2009, 284 (12), 7897−7902. (18) Hendrick, J. P.; Wickner, W. SecA protein needs both acidic phospholipids and SecY/E protein for functional high-affinity binding to the Escherichia coli plasma membrane. J. Biol. Chem. 1991, 266 (36), 24596−24600. (19) 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), 271−280. (20) Schaffitzel, C.; Ban, N. Generation of ribosome nascent chain complexes for structural and functional studies. J. Struct. Biol. 2007, 158 (3), 463−471. (21) Kuhn, P.; Weiche, B.; Sturm, L.; Sommer, E.; Drepper, F.; Warscheid, B.; Sourjik, V.; Koch, H.-G. The bacterial SRP receptor, SecA and the ribosome use overlapping binding sites on the SecY translocon. Traffic 2011, 12 (5), 563−578. (22) Sharma, S.; Davidson, A. L. Vanadate-induced trapping of nucleotides by purified maltose transport complex requires ATP hydrolysis. J. Bacteriol. 2000, 182 (23), 6570−6576. (23) Chapon, C. Role of the catabolite activator protein in the maltose regulon of Escherichia coli. J. Bacteriol. 1982, 150 (2), 722− 729. (24) Notley, L.; Ferenci, T. Differential expression of mal genes under cAMP and endogenous inducer control in nutrient-stressed Escherichia coli. Mol. Microbiol. 1995, 16 (1), 121−129. (25) Dean, D. A.; Reizer, J.; Nikaido, H.; Saier, M. H. Regulation of the maltose transport system of Escherichia coli by the glucose-specific enzyme III of the phosphoenolpyruvate-sugar phosphotransferase system. J. Biol. Chem. 1990, 265 (34), 21005−21010. (26) Panagiotidis, C. H.; Boos, W.; Shuman, H. A. The ATP-binding cassette subunit of the maltose transporter MalK antagonizes MalT, the activator of the Escherichia coli mal regulon. Mol. Microbiol. 1998, 30 (3), 535−546. (27) Säaf̈ , A.; Monné, M.; de Gier, J.-W.; von Heijne, G. Membrane topology of the 60-kDa Oxa1p homologue from Escherichia coli. J. Biol. Chem. 1998, 273 (46), 30415−30418. (28) Oliver, D. C.; Paetzel, M. Crystal structure of the major periplasmic domain of the bacterial membrane protein assembly facilitator YidC. J. Biol. Chem. 2008, 283 (8), 5208−5216. (29) Ravaud, S.; Stjepanovic, G.; Wild, K.; Sinning, I. The crystal structure of the periplasmic domain of the Escherichia coli membrane protein insertase YidC contains a substrate binding cleft. J. Biol. Chem. 2008, 283 (14), 9350−9358. (30) van Bloois, E.; Dekker, H. L.; Fröderberg, L.; Houben, E. N. G.; Urbanus, M. L.; de Koster, C. G.; de Gier, J.-W.; Luirink, J. 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 Lett. 2008, 582 (10), 1419−1424. (31) Rabilloud, T.; Chevallet, M.; Luche, S.; Lelong, C. Fully denaturing two-dimensional electrophoresis of membrane proteins: A critical update. Proteomics 2008, 8 (19), 3965−3973. (32) Barrera, N. P.; Robinson, C. V. Advances in the mass spectrometry of membrane proteins: From individual proteins to intact complexes. Annu. Rev. Biochem. 2011, 80 (1), 247−271. (33) Macher, B. A.; Yen, T.-Y. Proteins at membrane surfacesa review of approaches. Mol. BioSyst. 2007, 3 (10), 705−713. (34) Alami, M.; Dalal, K.; Lelj-Garolla, B.; Sligar, S. G.; Duong, F. Nanodiscs unravel the interaction between the SecYEG channel and its cytosolic partner SecA. EMBO J. 2007, 26 (8), 1995−2004. (35) Lim, K. H.; Madabhushi, S. R.; Mann, J.; Neelamegham, S.; Park, S. Disulfide trapping of protein complexes on the yeast surface. Biotechnol. Bioeng. 2010, 106 (1), 27−41. (36) Fischer, K. D.; Helms, J. B.; Zhao, L.; Wieland, F. T. Site-specific photocrosslinking to probe interactions of Arf1 with proteins involved in budding of COPI vesicles. Methods 2000, 20 (4), 455−464.

AUTHOR INFORMATION

Corresponding Author

*Phone: +1 604 822-5245. Fax: +1 604 822-5227. E-mail: [email protected].



ACKNOWLEDGMENTS We are thankful to Isabelle Kelly and Jenny Moon for technical assistance with mass spectrometry. This work was funded by the NSERC and the British Columbia Proteomics Network to F.D. and L.J.F. C.S.C. was supported by an NSERC postdoctoral fellowship. Laboratory infrastructure was provided by the Canada Foundation for Innovation, the BC Knowledge Development Fund and the BC Proteomics Network.



REFERENCES

(1) Wallin, E.; Heijne, G. V. Genome-wide analysis of integral membrane proteins from eubacterial, archaean, and eukaryotic organisms. Protein Sci. 1998, 7 (4), 1029−1038. (2) Weiner, J. H.; Li, L. Proteome of the Escherichia coli envelope and technological challenges in membrane proteome analysis. Biochim. Biophys. Acta 2008, 1778 (9), 1698−1713. (3) Reisinger, V.; Eichacker, L. A. Solubilization of membrane protein complexes for blue native PAGE. J. Proteomics 2008, 71 (3), 277−283. (4) Hooker, B. S.; Bigelow, D. J.; Lin, C. T. Methods for mapping of interaction networks involving membrane proteins. Biochem. Biophys. Res. Commun. 2007, 363 (3), 457−461. (5) Bernsel, A.; Daley, D. O. Exploring the inner membrane proteome of Escherichia coli: Which proteins are eluding detection and why? Trends Microbiol. 2009, 17 (10), 444−449. (6) Diaz-Mejia, J. J.; Babu, M.; Emili, A. Computational and experimental approaches to chart the Escherichia coli cell-envelopeassociated proteome and interactome. FEMS Microbiol. Rev. 2009, 33 (1), 66−97. (7) Tan, S.; Tan, H. T.; Chung, M. C. M. Membrane proteins and membrane proteomics. PROTEOMICS 2008, 8 (19), 3924−3932. (8) Bayburt, T. H.; Sligar, S. G. Membrane protein assembly into Nanodiscs. FEBS Lett. 2010, 584 (9), 1721−1727. (9) Ong, S.-E.; Blagoev, B.; Kratchmarova, I.; Kristensen, D. B.; Steen, H.; Pandey, A.; Mann, M. Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol. Cell. Proteomics 2002, 1 (5), 376−386. (10) Baba, T.; Ara, T.; Hasegawa, M.; Takai, Y.; Okumura, Y.; Baba, M.; Datsenko, K. A.; Tomita, M.; Wanner, B. L.; Mori, H. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol. Syst. Biol. 2006, 2, No. doi: 10.1038/msb4100050. (11) Dalal, K.; Duong, F. Reconstitution of the SecY translocon in nanodiscs. Methods Mol. Biol. 2010, 619, 145−156. (12) Foster, L. J.; de Hoog, C. L.; Mann, M. Unbiased quantitative proteomics of lipid rafts reveals high specificity for signaling factors. Proc. Natl. Acad. Sci. U. S. A. 2003, 100 (10), 5813−5818. (13) Rogers, L. D.; Kristensen, A. R.; Boyle, E. C.; Robinson, D. P.; Ly, R. T.; Finlay, B. B.; Foster, L. J. Identification of cognate host targets and specific ubiquitylation sites on the Salmonella SPI-1 effector SopB/SigD. J. Proteomics 2008, 71 (1), 97−108. (14) Mortensen, P.; Gouw, J. W.; Olsen, J. V.; Ong, S.-E.; Rigbolt, K. T. G.; Bunkenborg, J.; Cox, J. r.; Foster, L. J.; Heck, A. J. R.; Blagoev, B.; Andersen, J. S.; Mann, M. MSQuant, an open source platform for mass spectrometry-based quantitative proteomics. J. Proteome Res. 2009, 9 (1), 393−403. (15) Dalal, K.; Duong, F. The SecY complex: Conducting the orchestra of protein translocation. Trends Cell Biol. 2011, 9, 506−514. (16) Ishihama, Y.; Schmidt, T.; Rappsilber, J.; Mann, M.; Hartl, F. U.; Kerner, M.; Frishman, D. Protein abundance profiling of the Escherichia coli cytosol. BMC Genomics 2008, 9 (1), 102. (17) Dalal, K.; Nguyen, N.; Alami, M.; Tan, J.; Moraes, T. F.; Lee, W. C.; Maurus, R.; Sligar, S. S.; Brayer, G. D.; Duong, F. Structure, 1459

dx.doi.org/10.1021/pr200846y | J. Proteome Res. 2012, 11, 1454−1459