Journal of Microbiological Methods 98 (2014) 15–22

Contents lists available at ScienceDirect

Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth

Protein complex purification from Thermoplasma acidophilum using a phage display library Ágnes Hubert a, Yasuo Mitani b, Tomohiro Tamura c, Marius Boicu a, István Nagy a,⁎ a b c

Max-Planck Institute of Biochemistry, Department of Molecular Structural Biology, D-82152 Martinsried, Germany Bioproduction Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba 305-8566, Japan Bioproduction Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Sapporo 062-8517, Japan

a r t i c l e

i n f o

Article history: Received 29 October 2013 Received in revised form 12 December 2013 Accepted 12 December 2013 Available online 21 December 2013 Keywords: Protein complex Phage display library Thermoplasma acidophilum Electron microscopy

a b s t r a c t We developed a novel protein complex isolation method using a single-chain variable fragment (scFv) based phage display library in a two-step purification procedure. We adapted the antibody-based phage display technology which has been developed for single target proteins to a protein mixture containing about 300 proteins, mostly subunits of Thermoplasma acidophilum complexes. T. acidophilum protein specific phages were selected and corresponding scFvs were expressed in Escherichia coli. E. coli cell lysate containing the expressed Histagged scFv specific against one antigen protein and T. acidophilum crude cell lysate containing intact target protein complexes were mixed, incubated and subjected to protein purification using affinity and size exclusion chromatography steps. This method was confirmed to isolate intact particles of thermosome and proteasome suitable for electron microscopy analysis and provides a novel protein complex isolation strategy applicable to organisms where no genetic tools are available. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Optimal growth conditions of the thermo-acidophilic archaeon Thermoplasma acidophilum are around 59 °C and pH 2. Common feature of the genus Thermoplasma is that the cells lack a rigid cell wall and are delimited only by a plasma membrane. The genome size of T. acidophilum is small (1.5 Mbp) comprising 1507 open reading frames of which 1482 encode for proteins (Ruepp et al., 2000). The lack of cell wall, the relatively small genome, and the low cellular complexity make T. acidophilum a favorable model organism for visual proteomics approaches. This approach aims to determine spatial relationships of macromolecular complexes (over 300 kDa) inside an unperturbed cellular environment using cryo-electron tomography and pattern recognition procedures (Nickell et al., 2006). In the cytosolic proteome of T. acidophilum a large proportion of proteins is organized into multimeric complexes, amongst which 35 macromolecular assemblies have been identified with sizes over 300 kDa (Sun et al., 2009). These large protein structures are candidates for

Abbreviations: EM, electron microscopy; IMAC, immobilized metal affinity chromatography; MS/MS, tandem mass spectrometry; Ni–NTA, nickel–nitriloacetic acid; OHA-300, 300 mM K2HPO4–KH2PO4 elution fraction of Sup12-HMWF; OHA-500, 500 mM K2HPO4– KH2PO4 elution fraction of Sup12-HMWF; scFv, single-chain variable fragment; SEC, size exclusion chromatography; Sup12-HMWF, Superose12-separated high molecular weight protein fraction of T. acidophilum cytosolic extract; TAP, tandem affinity purification. ⁎ Corresponding author. Tel.: +49 89 8578 2648; fax: +49 89 8578 2641. E-mail addresses: [email protected] (Á. Hubert), [email protected] (Y. Mitani), [email protected] (T. Tamura), [email protected] (M. Boicu), [email protected] (I. Nagy). 0167-7012/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.mimet.2013.12.010

single particle analysis using electron microscopy (EM) to create template libraries toward visual proteomics studies and thereby promoting the generation of a comprehensive cellular atlas of macromolecular complexes. Proteomics studies based on 2DE-MALDI-TOF MS approach have provided information on the expressed cytosolic proteins and macromolecular complexes of T. acidophilum (Sun et al., 2007). In the protein complement of T. acidophilum there are a number of macromolecular assemblies playing important roles in protein folding, degradation, and metabolic pathways, including the archaeal chaperone thermosome (Nitsch et al., 1997), the VCP-like ATPase (VAT), which participates in numerous cellular activities (Gerega et al., 2005; Golbik et al., 1999; Pamnani et al., 1997; Rockel et al., 1999), the 20S proteasome (Zwickl et al., 1999) and the tricorn protease, the core of a modular proteolytic system (Tamura et al., 1996, 1998). Molecular sieve chromatography in combination with LC–MS/MS helped to reveal less abundant cytosolic proteins on the basis of size distribution (Sun et al., 2007). Until now, with the help of single particle electron microscopy and X-ray crystallography, many of these large complexes have been structurally characterized. However, there are still many hypothetical and partially characterized protein complexes whose molecular architecture and biochemical function is not yet explored. The characterization of these macromolecular assemblies could help to elucidate the structure and function of their similar but more complex eukaryotic homologues. To accelerate the purification of protein complexes novel methods have been developed. Chromosomally tagged protein purification technologies can be used if genetic tools are available for the host under investigation. Tandem affinity purification (TAP) tag is useful for rapid

16

Á. Hubert et al. / Journal of Microbiological Methods 98 (2014) 15–22

purification of complexes from a relatively small number of cells without prior knowledge of the complex composition, activity, or function. This tag can be integrated into the chromosome and TAP-tagged proteins can be expressed and purified using affinity chromatography. One of the key techniques of this method is that the tag sequence contains protease recognition sequence allowing proteolytic release of the bound material under native conditions (Rigaut et al., 1999). This method in combination with mass spectrometry enabled a large-scale approach to characterize multiprotein complexes in Saccharomyces cerevisiae, and provided an outline of the proteome as a network of protein complexes (Gavin et al., 2002). However, there are no genetic tools available for many archaea due to limitation in suitable selection markers, marker genes and transformation methods (Baka et al., 2013). Therefore alternative methods like the recently developed grid blotting technology using native polyacrylamide gel electrophoresis combined with EM grid blotting are being used to study protein complexes (Knispel et al., 2012). This is a fast and efficient method to transfer high molecular weight protein complexes from the acrylamide gel matrix directly to EM grids, facilitating structural analysis. Here we aimed at isolating native protein complexes from cytosolic extract of T. acidophilum using a combination of a phage display antibody library and a two-step chromatography method, thereby providing homogenous samples for EM-based structural studies. For this purpose we adapted the mono-antigen targeted phage display technology to a multi-antigen targeted version and generated a combinatorial scFvlibrary against the high molecular weight fraction of T. acidophilum, which contained more than 300 proteins as subunits of protein complexes with sizes larger than 300 kDa. Since protein complexes are sensitive to harsh purification conditions, our goal was to develop an antibody-based mild and fast protein isolation technique. The two-step chromatography method enabled protein complexes to keep their molecular assemblies in their intact and active form for subsequent structural and biochemical analyses.

2. Materials and methods 2.1. Strains and cell culture conditions T. acidophilum cells were cultured as described earlier with minor modifications (Sun et al., 2007; Robb, 1995). Briefly, 4 mL T. acidophilum cryo stock or 1 mL fresh culture was added to 50 mL medium and grown at 59 °C and 120 rpm in an oil bath until OD600 reached 1.0–1.2 (2–3 days). Cells were centrifuged at 4.000 ×g for 10 min at room temperature (RT), washed with distilled water, and stored at − 80 °C. Escherichia coli strains TG1 and HB2151 were purchased from Amersham Biosciences and strain BL21(DE3) was from Invitrogen. These strains were routinely cultured at 37 °C in LB (10 g L−1 tryptone, 5 g L−1 yeast extract, 10 g L−1 NaCl), 2× YT (16 g L−1 tryptone, 10 g L−1 yeast extract, 5 g L−1 NaCl) or YTG (16 g L−1 tryptone, 10 g L−1 yeast extract, 5 g L−1 NaCl, 2% (w/v) glucose) media amended with the appropriate antibiotics where needed.

2.2. Preparation of T. acidophilum cytosolic extract Frozen cell pellet was thawed on ice and resuspended in distilled water (2 mL for 1 g cell pellet) containing EDTA-free protease inhibitor cocktail (Roche). Cell lysis was triggered by elevating the pH of the suspension to 7.5 with non-buffered 1 M Tris. After cell lysis DNase I (Sigma-Aldrich) or Benzonase (Sigma-Aldrich) was added to digest released DNA and RNA and the cell lysate was incubated on ice for 1 h. Crude extract was centrifuged at 30.000 ×g and 4 °C for 45 min to remove cell debris and sediment the membrane fraction. Clear cytosolic extract was immediately used or frozen with 15% glycerol and stored at −20 °C.

2.3. Preparation of E. coli cytosolic extract Ten to fifteen grams of frozen cells were thawed on ice and resuspended in 20 mL milliQ water containing 1/2 tablet of EDTA-free protease inhibitor cocktail (Roche). The cell suspension was supplemented with 1 mg mL− 1 lysozyme and with the appropriate volume of 10 × Coupling buffer (0.08 M Na2HPO4, 0.02 M KH2PO4, 1.4 M NaCl, 0.1 M KCl, pH 7.4) to obtain 1 × buffer concentration and incubated on ice for 1 h. To decrease the viscosity of the lysate 50 μg mL− 1 of DNase (Sigma-Aldrich) or 2–3 μL of Benzonase (Sigma-Aldrich) was added. The lysate was sonicated on ice five to eight times for 1 min (duty cycle: 30%, output control: 6–8, Sonifier 250, Branson). Crude cell extract was centrifuged at 4 °C, 30.000 ×g, 40 min and the supernatant containing the soluble protein fraction was immediately used. 2.4. Column chromatography of macromolecular complexes Superose12 semi-preparative column (24 mL, GE-Healthcare) was used for the enrichment of macromolecular complexes from T. acidophilum cytosolic extract. The column was connected to an FPLC system (ÄKTA Purifier 10, GE Healthcare) and operated at 10 °C. Fivehundred microliters (max. 4 mg protein) of the cytosolic extract was loaded on the column that was equilibrated with running buffer (25 mM potassium phosphate buffer, pH 7.5 containing 1 mM ATP, 1 mM DTT and 5 mM MgCl2) and run at 0.4 mL min−1 flow rate. Protein elution profile was monitored with a UV-detector operated at 280 nm. Thirty-five protein-containing fractions of 0.6 mL were collected. High molecular weight protein complex (N 300 kDa) containing fractions including the void volume, were pooled and concentrated to 2.5 mL. This sample was mixed with 15% glycerol, aliquoted to 100–200 μL, frozen in liquid nitrogen, named as Sup12-HMWF and stored at − 80 °C. Sup 12-HMWF was used as antigen mixture for immunization of mice and for phage biopanning to obtain T. acidophilum phage library. The complexity of Sup12-HMWF was reduced by hydroxyapatite fractionation when it served as protein target mixture to capture specific phages. Sup12-HMWF was loaded on a 20 mL hydroxyapatite column (Bio-Gel HTP, Bio-Rad Laboratories) operated at 10 °C and equilibrated with loading buffer (10 mM potassium phosphate buffer, pH 8.0). Stepwise elution of bound proteins was carried out with 150, 300, and 500 mM potassium phosphate buffers (pH 8.0) at 0.5 mL min−1 flow rate. Fractions eluted with 300 and 500 mM potassium phosphate (OHA-300 and OHA-500, respectively) were dialyzed against 20 mM potassium phosphate buffer (pH 8.0) at 4 °C overnight (ON) and concentrated with an ultrafiltration membrane (5 kDa MWCO, Amicon Ultra, Merck Millipore). 2.5. Immunization of mice Eight Balb/c type 5–6 weeks old female mice were immunized three times with 45 μg of Sup12-HMWF protein mixture in 2 week intervals. The immunization procedure was carried out using Freund's adjuvant as described previously (Harlow and Lane, 1988). Spleens of mice were removed three days after the last injection and stored at −80 °C until use. Immunization and spleen removal of the animals were carried out according to the German Animal Protection Law by the authorized staff of the Animal House of MPI of Biochemistry. 2.6. Construction of scFv repertoire in pCANTAB5E phagemid vector Seventy mg mouse spleen was subjected to RNA extraction using the RNeasy Midi RNA extraction kit according to the manufacturer's protocol (Qiagen). RNA samples were recovered in 0.5 mL of DEPC-treated water and used immediately as template in RT-PCR reactions to synthesize cDNA. Remaining RNA was stored in 50 μL aliquots at −80 °C. The detailed protocol for library construction in pCANTAB5E phagemid vector (Amersham Pharmacia Biotech) is described in Short protocols in

Á. Hubert et al. / Journal of Microbiological Methods 98 (2014) 15–22

immunology (Coligan et al., 2005). The schematic work-flow of library construction is demonstrated in Fig. 1(A3–7). 2.7. Affinity selection of scFv-displaying phages on plastic-immobilized antigens For affinity selection scFv-displaying phages were propagated in E. coli TG1 (Amersham Biosciences) cells with the help of M13KO7 Helper phage (New England Biolabs) as described in Short protocols in immunology (Coligan et al., 2005). Three biopanning cycles were applied in each phage selection procedure. The ratio between panning input and output was monitored in each cycle. The schematic work-flow of phage biopanning is shown in Fig. 1(B1–6). 2.8. Screening of scFv-displaying phages and free scFvs using immunology methods

17

To monitor the specificity of His-tagged scFvs a sandwich of three antibodies (monoclonal His-tagged scFv as primary, His-probe (6–8) rabbit polyclonal IgG (Santa Cruz Biotechnology, sc-803) as secondary and goat anti-rabbit IgG-AP conjugate (Santa Cruz Biotechnology, sc2034) as tertiary antibody) was used. These antibodies were diluted 100–1000, 1000 and 5000 times in TBST, respectively. Monoclonal His-tagged scFv expression level was determined by His-probe (6–8) rabbit polyclonal IgG (Santa Cruz Biotechnology, sc803) as primary and goat anti-rabbit IgG-AP conjugate (Santa Cruz Biotechnology, sc-2034) as secondary antibodies according to the supplier recommendations. The One Step Western blot TMB solution (Thermo Scientific) was used for signal development in case of HRP-conjugated antibodies and the One Step Western blot BCIP/NBT solution (Thermo Scientific) or the Sigma Fast BCIP/NBT tablet (Sigma-Aldrich) was used to develop signals of alkaline phosphatase conjugated secondary antibodies. Phage screening steps are shown in Fig. 1(B 7–9). 2.9. Expression of His-tagged scFv

Phages from single infected bacterial colonies were propagated in E. coli TG1 cells with the help of M13KO7 Helper phage and screened by monoclonal Enzyme-Linked Immunosorbent Assay (ELISA) to determine antigen specificity. Individual TG1 colonies were picked from library output plates and inoculated into 100 μL YTCbG medium (YTG amended with 100 μg mL−1 carbenicillin) in a 96-well flat bottom microtiter plate (Corning) and cells were grown ON at 37 °C with shaking at 600 rpm. Next day, 10 μl cell suspensions were inoculated into 100 μL fresh YTCbG medium in microtiter plates and grown at 37 °C, 600 rpm on an Eppendorf Thermomixer for 1 h (until OD600 reached ~0.5–0.7). At this point 25 μl (109 PFU) M13KO7 helper phage suspension (in YTCbG medium) was added to each well and the plate was incubated at 37 °C for 30 min at first without, then with shaking for 1 h at 600 rpm. After incubation the plate was centrifuged at 1300 ×g for 15 min at RT. Cells were resuspended in 200 μl 2 × YT medium amended with 100 μg mL−1 carbenicillin and 50 μg mL−1 kanamycin after gentle removal of supernatant and grown at 30 °C, 700 rpm ON. Cell debris was removed by centrifugation at RT, 1300 ×g for 30 min next morning and released phages in the supernatant were transferred to ELISA-plates (MaxiSorp, Nunc). Prior to phage testing ELISA plates were coated with 100 μL/well of either Sup12-HMWF or OHA 500 target protein solutions (10–100 μg mL− 1) diluted in PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4·2 H2O, 2 mM KH2PO4) and incubated ON at 4 °C with gentle shaking. A negative control plate incubated with 100 μL/well MPBS (2% skimmed milk in PBS) was prepared for each antigen plate. Wells were washed three times with 350 μL PBS using an automated microtiter plate-washer (Biotek ELx50/16). To block remaining binding surfaces, wells were filled with 250 μL MPBS and incubated with shaking at RT for 2 h. Plates were washed three times with PBS then 100 μL PBST (PBS containing 0.1% Tween 20) was added to each well to which 90 μL of the corresponding phage suspension was added. Plates were covered with parafilm and incubated at RT for 1.5 h. Wells were washed three times with PBST then three times with PBS. For phage detection HRP-conjugated anti-M13KO7 antibody (Amersham Pharmacia Biotech) or HRP-conjugated anti-E-tag antibody (Bethyl Laboratories) were used. Plates were incubated with the antibody solution at RT for 1 h then washed three times with PBST and three times with PBS. To develop ELISA-signals 5-aminosalicylic acid (Sigma) substrate solution or Ultra TMB-ELISA substrate solution (Thermo Scientific) was used. To determine phage specificity Western blot analyses were carried out. A horizontal, semi-dry electro-blotting system (Bio-rad) was used for transferring proteins to nitrocellulose membrane (Whatman) according to the supplier recommendations. Remaining binding surfaces were blocked with 3% MPBS. ScFv-displaying phages were diluted in PBST (5 ×) and served as primary antibodies. A 5000 × dilution of Mouse anti-M13-HRP conjugate (Amersham Pharmacia Biotech) in PBST was used as secondary antibody.

To clone scFv genes from pCANTAB5E phagemid vectors (Amersham Pharmacia Biotech) into pET28a expression vector (Novagen), an SfiI restriction site was introduced into the pET28 vector in a PCR reaction using primers 5′-CCGCAAGGAATGGTGCATGCAAGGAGATGG3′ and 5′-GGTACGATGGCCGGCTGGGCCATGGTATATCTCCTTCTTAA AGTTAAA-3′. The PCR reaction was carried out with the following program: 95 °C for 5 min, 35 cycles of 95 °C for 30 s, 55 °C for 1 min, 72 °C for 1 min, and followed by 72 °C for 5 min. Miniprep DNA of vector pET28a was used as template. The obtained DNA fragment was digested with SphI and ligated into pET28a digested with SphI and Ecl136II restriction endonucleases. The resulting plasmid contained the SfiI site adjacent to the NcoI site and was designated as pET28(SfiI). Phagemid SfiI/NotI fragment encoding for the scFv gene was ligated into the corresponding positions of pET28(SfiI) having the 6× His-tag at the C terminus. E. coli BL21(DE3) cells were transformed and single colonies were inoculated into 3–100 mL kanamycin (50 mg mL−1) containing media and grown ON at 37 °C. Bacteria were then transferred into fresh LB-medium and incubated at 37 °C and 200 rpm in the presence of the appropriate antibiotic. When OD600 0.8 was reached 1 mM IPTG was added to induce protein expression and the cell culture was further incubated at 37 °C for 5 h. 2.10. Two-step chromatography Cell lysates from specific His-tagged scFv harboring E. coli BL21(DE3) and T. acidophilum cells were prepared in 1× Coupling buffer. Crude cell extracts were mixed and sedimented at 30.000 ×g and 4 °C for 40 min. The supernatant was transferred into a new polypropylene tube and incubated with continuous rotation at RT for 1 h. The sample was supplemented with 10 mM imidazol and loaded on a 1 mL HisTrap FF crude immobilized metal ion affinity chromatography (IMAC) column (GE Healthcare Life Sciences), which was connected to an ÄKTA purifier FPLC system. The purification was carried out at 10 °C in a cold cabinet with a flow rate of 0.8 mL min−1. The elution profile was monitored with a UV-detector operated at 230, 254, and 280 nm. Prior to sample loading the column was equilibrated with running buffer (50 mM sodium phosphate buffer, pH 8.0 and 300 mM NaCl and 10 mM imidazole). Unbound and weakly bound proteins were washed off with 20 mM imidazole-containing running buffer until A280 reached a constant base line. Bound proteins were eluted with a 20–250 mM imidazole gradient, which was expanded to 40–45 column volumes. Fractions of 0.8 mL were collected and analyzed by SDS-PAGE. Fractions of interest were pooled and concentrated by an ultrafiltration membrane (Amicon Ultra centrifugal filter unit, Ultra-15, MWCO 30 kDa, Millipore) or directly used.

18

Á. Hubert et al. / Journal of Microbiological Methods 98 (2014) 15–22

Fig. 1. Work-flow of phage display library construction and two-step chromatography purification of T. acidophilum complexes demonstrated on the example of thermosome. High molecular weight protein fractions of T. acidophilum (red box) were separated on a Superose12 column (Sup12-HMWF) (A1) and eight Balb/c type 5–6 weeks old female mice were immunized three times with 45 μg of Sup12-HMWF protein mixture in 2 week intervals (A2). Spleens of mice were removed three days after the last injection and stored at −80 °C until use (A3). RNA was extracted from 70 mg mouse spleen (A4) and used as template for cDNA synthesis (A5). scFvs were synthesized in consecutive PCR reactions. First the variable regions of heavy (VH) and light (VL) chains were amplified and SfiI and the NotI sites were introduced, respectively, followed by the introduction of the linker region and scFv amplification (A6). scFvs were cloned into the pCANTAB5E phagemid vector (A7). T. acidophilum specific scFv displaying phages were affinity selected/enriched using Sup12-HMWF in biopanning experiments (B1–6) and a sublibrary was also prepared against proteins of the OHA-500 (sub-fraction of Sup12-HMWF) in biopanning experiments. The phage display library was propagated (B1) and purified (B2). Target proteins of Sup12-HMWF (or OHA-500) were attached to test tubes and incubated with phages (input phages) (B3). Unbound phages were eliminated via washing steps (B4), bound phages (output phages) were eluted from target proteins (B5) and TG1 cells were infected with them (B6); monoclonal phagemids were plated after E. coli transfection (B7) and antigen specificity of monoclonal scFv-displaying phages was determined by Enzyme-Linked Immunosorbent Assay (ELISA) against OHA-500 proteins (B8). Western blot analysis was carried out to reveal the MW of the target protein and scFv specificity (B9). Specific scFv genes were cloned into pET28 (SfiI) vector (B10) to obtain higher yield and soluble His-tagged scFvs were expressed in E. coli cells (B11). T. acidophilum cells harboring prey complexes (C1) and E. coli cells harboring bait His-scFvs (C2) were lysed. Cell lysates were mixed (C3) and cell debris was removed by centrifugation at 30,000 xg for 40 min (C4). Binding of the bait and prey was promoted by gentle shaking on a rotating platform for 1.5 h at RT (C5). Thermosomes were captured and bound to Ni2+ Sepharose resin via His-tagged scFvs separating them from other proteins in the flow-through fraction (C6). Affinity bound proteins were eluted with imidazole gradient and target protein fractions (indicated with red rectangle) were pooled (C7), concentrated and subjected to size exclusion chromatography (SEC) using a Superose6 column (C8) for further separation of the bait–prey complex from contaminants. Fractions of SEC separated proteins (C9) were analyzed by SDS-PAGE and MS/MS and an aliquot of the fractions/ proteins of interest (His-scFv-themosome complexes indicated in the red box) was put on an EM grid (C10) and analyzed by EM. Images were recorded for single particle analyses (C11).

Size exclusion chromatography was used for further purification of scFv-captured macromolecular complexes. A 24 mL Superose6 column (GE-Healthcare) was connected to ÄKTA

Purifier 10 (GE Healthcare) and operated at 0.4 mL min − 1 flow rate and 10 °C. The protein elution profile was monitored with a UV-detector operated at 280 nm. Samples of 300–500 μL were

Á. Hubert et al. / Journal of Microbiological Methods 98 (2014) 15–22

19

loaded and 0.4 mL fractions were collected. Sample purity was analyzed by SDS-PAGE. 2.11. Mass spectrometry Protein samples of interest were digested with trypsin either in solution or in gel for tandem mass spectrometry (MS/MS) analysis (Shevchenko et al., 1996). Digested peptide mixtures were separated by on-line nanoLC and analyzed by electronspray MS/MS. The experiments were performed on an Agilent 1200 nanoflow system connected to an LTQ Orbitrap mass spectrometer (Thermo Electron, Bremen, Germany) equipped with a nanoelectrospray ion source (Proxeon Biosystems, Odense, Denmark). Data analysis was performed with the MaxQuant software as described (Gruhler et al., 2005) supported by Mascot or Andromeda as database search engines for peptide identifications. Peaks in MS scans were determined as three-dimensional hills in the mass-retention time plane. MS/MS peak lists were filtered to contain at most six peaks per 100 Da interval and searched against a concatenated forward and reversed version of the T. acidophilum and E. coli databases (extracted from NCBInr) and Mouse database (International Protein Index database: ftp://ftp.ebi.ac.uk/pub/databases/IPI/old/MOUSE) and frequently observed contaminants like proteases and human keratins. The initial mass tolerance in MS mode was set to 7 ppm and MS/MS mass tolerance was 0.5 Da. Cysteine carbamidomethylation was searched as a fixed modification, whereas N-acetyl protein and oxidized methionine were searched as variable modifications. 2.12. Electron microscopy Isolated protein complexes were negatively stained and subjected to single particle EM-analysis. Copper grids (100 × 400 meshes coated with a continuous carbon film, PLANO GmbH) were glow discharged for 30 s in a plasma-cleaner, after which they were placed C-side down on a 5 μL drop of the protein sample for 60 s and blotted with a filter paper (Grade No. 1, Whatman). In the same way, grids were washed and blotted quickly 2 times with 5 μL milliQ water and incubated for 30 s on a 2 μL droplet of 2% uranyl-acetate before drying by blotting with filter paper. Electron micrographs were acquired with a transmission electron microscope (CM 200 FEG TEM, 160 kV, Philips) equipped with a TVIPS CCD camera (CCD size: 4096 × 4096 pixels, CCD physical pixel size: 15 mue, conversion factor: 67 counts per incident electron). The applied defocus and magnification of the images varied between −2 and −3 μm and 539,600 and 93,126, respectively. 3. Results 3.1. Creation of scFv library and phage selection against antigen mixtures Proteins of Sup12-HMWF were used to immunize mice. Spleens of the animals were harvested after three times immunization and subjected to RNA isolation/scFv library construction. The scFv-library contained 1.4 × 108 clones of which 6 clones were randomly selected and sequenced. All the sequences were different indicating a diverse scFv library (data not shown). To enrich T. acidophilum specific phages we used the Sup12-HMWF protein fraction in biopanning experiments. After three biopanning cycles at least 15 different protein bands were observed by Western blot analysis providing evidence for the presence of specific scFv-binders (Fig. 2). Sub-fractions OHA-500 and OHA-300 of Sup12-HMWF were used as baits to select individual scFv-displaying phages through three biopanning cycles. These protein fractions have been described to contain well-known protein complexes (Knispel et al., 2012). In case of OHA-500 the output phage number increased to 2 × 10 7 CFU mL− 1 while after selection on OHA-300 the phage titer remained 1 × 103 CFU mL−1. These results indicated the presence of bait proteins in fraction OHA-500.

Fig. 2. Western blot assays of output 3 phages selected against Sup12-HMWF (A) and OHA-500 (B) protein mixtures. The applied test antigens blotted on the membrane were Sup12-HMWF (1), T. acidophilum cytosolic extract (2) and OHA-500 (3). M: molecular weight marker.

Output phage samples of OH-500 and OH-300 were amplified and used as polyclonal phage antibody sources in Western blot analyses to monitor phage binding ability to denatured proteins of corresponding fractions. Two major protein bands at ~58 and ~26 kDa were visualized in sample OH-500 (Fig. 2) and as it was expected, no positive binders were obtained for OHA-300 (data not shown). E. coli TG1 cells were infected with output phages and single colonies were isolated as monoclonal-phage producing hosts. Monoclonal phages were screened against native proteins of OHA-500 by ELISA. More than 50% of the screened phages gave positive signal (data not shown). Binding phages were further screened by Western blot assay. Twenty-four ELISA-positive monoclonal phages were tested for protein binding properties, of which seven gave positive signal for the ~58 kDa and three for the ~ 26 kDa protein, corresponding to the recognition pattern of polyclonal phages. Additionally, one clone gave a weak signal at ~20 kDa (data not shown).

3.2. Expression of recombinant scFvs derived from positive phagemid clones To obtain scFv protein, we first used E. coli HB2151 non-suppressor strain which allows cytosolic, periplasmic and partially extracellular E-tagged scFv expression from the phagemid without the need to reclone scFvs into a new expression vector. The monoclonal ELISA test of soluble E-tagged scFvs resulted in weak positive signal intensities compared to those expressed on phages, indicating that the concentration of free-form scFvs in the medium is significantly lower than their displayed versions. Next, promising scFv inserts were re-cloned into a modified pET28 vector (pET28[SfiI]) to allow C-terminal 6× His-tagged scFv expression in the cytosol. Soluble scFv expressions were assayed by Western blot using anti-His tag antibody. Eight scFv candidates were subjected to small-scale affinity-purification using Ni–NTA beads. Western blot assays of the eluted fractions provided evidence for soluble scFv expression with varying intensities (Fig. 3).

20

Á. Hubert et al. / Journal of Microbiological Methods 98 (2014) 15–22

Fig. 3. Western blot assay of Ni–NTA-purified soluble 6× His-scFvs. Lanes from C2 to H10 represent affinity-purified 6× His scFvs. The variation in size (27–29 kDa) of His-tagged scFvs is due to sequence length differences in variable segments of heavy and light chains. M: molecular weight marker.

3.3. Purification and EM analysis of protein complexes Protein complexes were captured by His-tagged scFvs and purified in a two-step chromatography experiment (Fig. 1C). E. coli cell extract containing the expressed scFv (bait) and T. acidophilum cell lysate containing target protein complexes (pray) were mixed and incubated, after which the mixture was subjected to Ni–NTA-based affinity chromatography. A size exclusion chromatography step was introduced into the purification process to further separate the bait/pray complex

from contaminants and to get information on protein sizes. This method did not require purified scFvs and enabled the use of mild buffer conditions (pH, salt components and concentration) during purification steps. For proof of concept the ~ 58 kDa protein band recognizing scFv (scFv-C8) was subjected to this chromatography method. IMAC fractions containing the scFv/protein complex were pooled and subjected to Superose6 chromatography. The peak maximum was at 11.5 mL retention volume (fractions 14–15), which corresponded to ~ 900– 1000 kDa molecular weight (Fig. 4). The MS/MS analysis of pooled fractions (11–17) identified three proteins Ta0980 (α) and Ta1276 (β) subunits of the molecular chaperon thermosome and the scFv (28.46 kDa). This purification yielded 80 μg thermosomes from 2 g T. acidophilum biomass with 98% purity. Single particle electron microscopy analysis in negative stain showed homogenous particle set perfect for data collection, class averaging and 3D reconstruction (Fig. 5). The side and top views of the complex show the stacking of two “eight-member ring” architecture. A similar protein purification experiment was carried out with the ~26 kDa protein recognizing scFv (scFv-A11). The first elution peak of the size exclusion chromatography was at 12.5 mL retention volume (fractions 17–18), which corresponded to 600–800 kDa MW (Fig. 4). scFv-containing protein fractions were analyzed by MS/MS and by single particle EM in negative stain. The MS/MS analysis identified α and β subunits of T. acidophilum proteasome, Ta1288 and Ta0612, respectively and the scFv. Single particle EM-analysis of pooled fractions revealed the characteristic shape of side views of proteasome particles (Fig. 5). Finally, we purified the ~ 20 kDa/scFv-G11 protein complex. The chromatography peak maximum was obtained at 15.4 mL retention volume, corresponding to a molecular weight of 150–250 kDa. The MS/MS analysis identified protein Ta0152 a probable peroxiredoxin. Due to this small size the EM analysis was not successful (data not shown).

Fig. 4. SDS-PAGE (15%) analysis of scFv-captured thermosomes (A) and proteasomes (B) separated on Superose6 SEC column. Subunits of the thermosome (58.28 kDa and 58.48 kDa) are present in fractions 11–17 together with the scFv-C8 (28.46 kDa). Subunits of the proteasome (25.8 kDa and 23.2 kDa) are present in fractions 16–18 together with the scFv-A11 (28 kDa) M: molecular weight marker. Numbers in the upper row indicate fraction numbers.

Á. Hubert et al. / Journal of Microbiological Methods 98 (2014) 15–22

21

Fig. 5. Electron micrographs of scFv-captured, Superose6-purified and negatively stained thermosome (A) and proteasome (B) particles. The white scale bar corresponds to 100 nm. A1– B1: class average of side view, A2: class average of top view, A3, B2: 3D-reconstruction from the data.

4. Discussion In this study, we obtained monoclonal scFv-displaying phages of which scFv-C8, scFv-A11 and scFv-G11 recognized thermosome, proteasome and peroxiredoxin complexes, respectively. Out of 24 ELISApositive clones, 11 clones recognized one of these proteins in Western blot analyses. The other 13 clones did not recognize linear epitopes of denatured proteins subject to Western analysis most probably they recognize tertiary or quaternary structures of native proteins. The abundance of thermosomes and proteasomes and the peroxiredoxin in the applied antigen mixture favored their corresponding scFvs to be selected during the biopanning procedure. Thus, scFv-displaying phages which were specific to protein complexes with low expression levels escaped capturing. It is also possible that the original phage library contained limited species of scFvs although it is not likely as the Western analysis of Sup12-HMWF showed many more signals than OHA500 (Fig. 2). To enrich phage specific for poorly expressed protein complexes, the removal of scFvs against abundant proteins with affinity selection using heterologously expressed proteins seems an obvious option. The 20S proteasome of T. acidophilum is interacting with the CDC48 homolog ATPase (Ta0840) (Barthelme and Sauer, 2012; Forouzan et al., 2012). This interaction was demonstrated with recombinant proteins in pull-down and biochemical assays. However, Ta0840 was not detected in our pull-down assays using two distinct 20S proteasome specific scFv antibodies. This might be due to the different compositions of lysis buffer that was used in the above mentioned experiments. This supports the observation that to isolate loosely associated complexes in intact shape structure-preserving buffer compositions should be determined experimentally for each complex, regardless of the protein separation method used to isolate them (Sun et al., 2009). Previous proteomics studies showed that Ta0152 possesses complex forming ability suggesting a ~500 kDa molecular weight, which size was in good agreement with its hexadecameric homologue of Aeropyrum pernix (Mizohata et al., 2005). Our result might indicate that the antibody purified Ta0152 ring was prevented from ordered aggregation which was detected in earlier studies (R. Knispel, personal communication) therefore this complex could not be visualized by EM (data not shown). Our scFv-antibody based two-step chromatography separation method allowed the isolation of intact complexes from native T. acidophilum cell extract and on the example of two well-known complexes we

demonstrated the feasibility of scFv-based purification technique for other high molecular weight complexes. These complexes can be subjected to biochemical and/or EM analyses however care should be taken to carry out the experiments with several specific scFvs to avoid enzyme inhibition. We proved the power of this method by described experiments however to avoid unnecessary cloning steps and to increase the bait yield creating a His-tagged nanobody displaying phage library based on camelid immunization would be a viable option. Acknowledgments We thank the crew of MPI of Biochemistry Core facility and the Animal House for their contribution to the project, especially Dr. Sabine Suppmann for recombinant protein expression, Snezan Marinkovic for DNA-sequencing and Dr. Cyril Boulegue for performing the LC–MS/ MS-analyses of protein samples. We also thank Dr. Parijat Majumder (MPI of Biochemistry) for her contribution to figure preparation. References Baka, E., Varga, S., Hobel, C., Knispel, R.W., Fekete, C., Ivanics, M., Kriszt, B., Nagy, I., Kukolya, J., 2013. The first transformation method for the thermo-acidophilic archaeon Thermoplasma acidophilum. J. Microbiol. Methods 95, 145–148. Barthelme, D., Sauer, R.T., 2012. Identification of the Cdc48*20S proteasome as an ancient AAA+ proteolytic machine. Science 337, 843–846. Coligan, J.E., Bierer, B., Margulies, D., Shevach, E., Strober, W., 2005. Short Protocols in Immunology. Wiley. Forouzan, D., Ammelburg, M., Hobel, C.F., Stroh, L.J., Sessler, N., Martin, J., Lupas, A.N., 2012. The archaeal proteasome is regulated by a network of AAA ATPases. J. Biol. Chem. 287, 39254–39262. Gavin, A.C., Bosche, M., Krause, R., Grandi, P., Marzioch, M., Bauer, A., Schultz, J., Rick, J.M., Michon, A.M., Cruciat, C.M., Remor, M., Hofert, C., Schelder, M., Brajenovic, M., Ruffner, H., Merino, A., Klein, K., Hudak, M., Dickson, D., Rudi, T., Gnau, V., Bauch, A., Bastuck, S., Huhse, B., Leutwein, C., Heurtier, M.A., Copley, R.R., Edelmann, A., Querfurth, E., Rybin, V., Drewes, G., Raida, M., Bouwmeester, T., Bork, P., Seraphin, B., Kuster, B., Neubauer, G., Superti-Furga, G., 2002. Functional organization of the yeast proteome by systematic analysis of protein complexes. Nature 415, 141–147. Gerega, A., Rockel, B., Peters, J., Tamura, T., Baumeister, W., Zwickl, P., 2005. VAT, the thermoplasma homolog of mammalian p97/VCP, is an N domain-regulated protein unfoldase. J. Biol. Chem. 280, 42856–42862. Golbik, R., Lupas, A.N., Koretke, K.K., Baumeister, W., Peters, J., 1999. The Janus face of the archaeal Cdc48/p97 homologue VAT: protein folding versus unfolding. Biol. Chem. 380, 1049–1062. Gruhler, A., Olsen, J.V., Mohammed, S., Mortensen, P., Faergeman, N.J., Mann, M., Jensen, O.N., 2005. Quantitative phosphoproteomics applied to the yeast pheromone signaling pathway. Mol. Cell. Proteomics 4, 310–327. Harlow, E., Lane, D., 1988. Antibodies: A Laboratory Manual. CSHL Press.

22

Á. Hubert et al. / Journal of Microbiological Methods 98 (2014) 15–22

Knispel, R.W., Kofler, C., Boicu, M., Baumeister, W., Nickell, S., 2012. Blotting protein complexes from native gels to electron microscopy grids. Nat. Methods 9, 182–184. Mizohata, E., Sakai, H., Fusatomi, E., Terada, T., Murayama, K., Shirouzu, M., Yokoyama, S., 2005. Crystal structure of an archaeal peroxiredoxin from the aerobic hyperthermophilic crenarchaeon Aeropyrum pernix K1. J. Mol. Biol. 354, 317–329. Nickell, S., Kofler, C., Leis, A.P., Baumeister, W., 2006. A visual approach to proteomics. Nat. Rev. Mol. Cell Biol. 7, 225–230. Nitsch, M., Klumpp, M., Lupas, A., Baumeister, W., 1997. The thermosome: alternating alpha and beta-subunits within the chaperonin of the archaeon Thermoplasma acidophilum. J. Mol. Biol. 267, 142–149. Pamnani, V., Tamura, T., Lupas, A., Peters, J., Cejka, Z., Ashraf, W., Baumeister, W., 1997. Cloning, sequencing and expression of VAT, a CDC48/p97 ATPase homologue from the archaeon Thermoplasma acidophilum. FEBS Lett. 404, 263–268. Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M., Seraphin, B., 1999. A generic protein purification method for protein complex characterization and proteome exploration. Nat. Biotechnol. 17, 1030–1032. Robb, F.T., 1995. Archaea: A Laboratory Manual. Cold Spring Harbor Laboratory Press. Rockel, B., Walz, J., Hegerl, R., Peters, J., Typke, D., Baumeister, W., 1999. Structure of VAT, a CDC48/p97 ATPase homologue from the archaeon Thermoplasma acidophilum as studied by electron tomography. FEBS Lett. 451, 27–32.

Ruepp, A., Graml, W., Santos-Martinez, M.L., Koretke, K.K., Volker, C., Mewes, H.W., Frishman, D., Stocker, S., Lupas, A.N., Baumeister, W., 2000. The genome sequence of the thermoacidophilic scavenger Thermoplasma acidophilum. Nature 407, 508–513. Shevchenko, A., Wilm, M., Vorm, O., Mann, M., 1996. Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal. Chem. 68, 850–858. Sun, N., Beck, F., Knispel, R.W., Siedler, F., Scheffer, B., Nickell, S., Baumeister, W., Nagy, I., 2007. Proteomics analysis of Thermoplasma acidophilum with a focus on protein complexes. Mol. Cell. Proteomics 6, 492–502. Sun, N., Tamura, N., Tamura, T., Knispel, R.W., Hrabe, T., Kofler, C., Nickell, S., Nagy, I., 2009. Size distribution of native cytosolic proteins of Thermoplasma acidophilum. Proteomics 9, 3783–3786. Tamura, T., Tamura, N., Cejka, Z., Hegerl, R., Lottspeich, F., Baumeister, W., 1996. Tricorn protease—the core of a modular proteolytic system. Science 274, 1385–1389. Tamura, N., Lottspeich, F., Baumeister, W., Tamura, T., 1998. The role of tricorn protease and its aminopeptidase-interacting factors in cellular protein degradation. Cell 95, 637–648. Zwickl, P., Voges, D., Baumeister, W., 1999. The proteasome: a macromolecular assembly designed for controlled proteolysis. Philos. Trans. R. Soc. Lond. B Biol. Sci. 354, 1501–1511.

Protein complex purification from Thermoplasma acidophilum using a phage display library.

We developed a novel protein complex isolation method using a single-chain variable fragment (scFv) based phage display library in a two-step purifica...
1MB Sizes 0 Downloads 0 Views