Protocol

A Combined Approach for the Localization and Tandem Affinity Purification of Protein Complexes from Metazoans

See allHide authors and affiliations

Science's STKE  11 Jan 2005:
Vol. 2005, Issue 266, pp. pl1
DOI: 10.1126/stke.2662005pl1

Abstract

An understanding of a protein's function is greatly aided by the knowledge of its localization in vivo and identification of any interacting partners. Here, we describe a method, based on expression of a genetically encoded fusion protein, that allows for protein localization and affinity purification (LAP) in metazoans. This method makes it possible to rapidly identify transformants and to conduct live imaging of protein localization and dynamics. In addition, the same tag can be used in a modified tandem affinity purification (TAP) procedure to isolate native protein complexes of high purity. The efficacy of this purification procedure allows for the characterization of rare protein complexes that are largely inaccessible to classical biochemical purifications. The LAP strategy should be widely applicable to the characterization of protein function in cellular and developmental contexts in various metazoans.

Introduction

Recent progress toward defining the complete proteomes of various eukaryotic organisms has generated a need for efficient procedures that determine the functions of newly identified proteins. A determination of the subcellular environment to which a protein localizes provides an important clue to protein function. In addition, the majority of proteins within a cell are present in complexes containing several polypeptides. For example, of the 589 proteins isolated using high-throughput purifications from budding yeast, 78% copurified with at least one additional protein (1). Defining the physical interactions that exist between proteins in a cell can help assign novel proteins to specific functional classes and elucidate the protein networks that contribute to specific biological processes.

The discovery of fluorescent proteins from jellyfish and corals that are functional in living cells, such as green fluorescent protein (GFP) and DsRed (and its monomeric counterpart mRFP), has greatly facilitated the analysis of dynamic localization of fusion proteins [reviewed in (2)]. Similarly, the use of epitope tags has provided generalized methods for protein purification. In particular, the development of the tandem affinity purification (TAP) tag by Séraphin and co-workers (3) has provided a means of cleanly isolating native protein complexes, even those present in limiting amounts in a cell.

The principle behind the TAP tag approach is to express a genetically encoded fusion protein with two distinct purification tags, which have a specific protease cleavage site between them (3). The fusion protein is isolated by the binding of the first tag to an appropriate affinity matrix. This tag is then removed by cleavage with a specific protease releasing the remaining fusion protein into the supernatant. A second round of purification, which relies on the second tag, further enriches for the fusion protein. The original TAP tag utilized the minimal domain of protein A (its Z domain; binds to IgG) and calmodulin-binding peptide (CBP; binds to calmodulin), with the highly specific tobacco etch virus (TEV) protease site between them. However, numerous other variations have been constructed based on other purification tags, including 6×His (binds to nickel), S peptide (binds to S protein), hemagglutinin (HA) peptide (recognized by antibodies to HA), and myc peptide (recognized by antibodies to myc), as well as other site-specific proteases, including Amersham’s PreScission protease 3C (46).

TAP tag procedures have been utilized primarily in budding and fission yeast, in which homologous recombination allows the efficient integration of the tag at the C terminus of the genomic locus in a haploid strain. This integration permits the expression of the fusion protein under the endogenous promoter, and provides a built-in ability to determine the functionality of the fusion, because it is the only copy of the gene in these haploid unicellular eukaryotes. Recently, two collections of yeast strains have been constructed in which virtually the complete set of open reading frames was fused to either GFP or the TAP tag (7, 8). Although methods exist to transform many other eukaryotes, the lower efficiency of homologous recombination (compared to yeast) has prevented the type of high-throughput molecular replacements required to effectively utilize the TAP tagging strategy.

To develop procedures that would facilitate the analysis of protein complexes in metazoans, including Caenorhabditis elegans and humans, we designed a modified version of the TAP tag (Fig. 1) that can be additionally used for localizing proteins in living cells using GFP. In our purification procedure, GFP is also used as the first purification tag (instead of the Z domain of protein A in the original TAP tag). Because this type of genetic tag can be used for both localization and affinity purification, we refer to it as the "localization and affinity purification" (LAP) tag. The use of GFP as the primary purification tag has two additional advantages. First, transformed cells or strains can been identified by screening for fluorescence, and cells or strains expressing an appropriate level of the fusion protein can be selected using either fluorescence microscopy or flow cytometry. Second, the localization dynamics of the expressed fusion protein can serve as a partial indicator of its functionality. For example, if the endogenous protein localizes to mitotic chromosomes, then the expressed LAP fusion should also localize to mitotic chromosomes. Although this is not equivalent to genetic rescue in haploid yeast cells, it offers some indication of whether the exogenously expressed protein retains the functionality of the endogenous protein. Here, we present a generalized strategy to purify native protein complexes using the LAP tag. Although the Protocol described here focuses on the use of this tag in cultured human cells, we have successfully used this strategy in both tissue culture cells and C. elegans (9). This approach should be applicable to any organism in which it is possible to generate stable transformants.

Fig. 1.

Overview of the LAP purification method.

Materials

Cell Culture

6-well culture plates

10-cm and 15-cm tissue culture dishes, tissue culture-treated by vacuum gas plasma, Falcon

Dimethyl sulfoxide (DMSO), sterile (Sigma-Aldrich, St. Louis, MO, #D-2650)

Dulbecco’s phosphate-buffered saline [Dulbecco’s PBS; Gibco, #14190-144 (http://www.invitrogen.com/gibco)]

Dulbecco’s modified eagle medium (DMEM; Gibco, #11960)

Eight-chambered polystyrene Lab-Tek coverglass system [Nunc, #155411 (http://www.nuncbrand.com)]

Ethanol

Heat-inactivated fetal bovine serum (FBS; Gibco, #16000-036)

L-Glutamine, 100× (Gibco, #25030-149)

Penicillin-streptomycin, 100×(10,000 units/ml of penicillin and 10,000 μg/ml of streptomycin; Gibco, #15140-148)

0.05% Trypsin-ethylene diamine tetra-acetic acid (EDTA) (Gibco, #25300-054)

Chemicals

2-mercaptoethanol(2-ME)

Affi-Prep protein A beads (BioRad, Hercules, CA, #156-0006)

Affinity-purified GFP antibodies

Note: For our purifications, we generated a polyclonal rabbit anti-GFP antibody through Covance Research Products, and affinity-purified it using standard procedures(10). Affinity-purified anti-GFP antibodies are available from various sources, including Covance [#MMS-118P (http://www.covance.com)], Santa Cruz Biotechnology (Santa Cruz, CA, #sc-8334), and Research Diagnostics (Flanders, NJ, #RDI-GRNFP3abm). In addition, antibodies precoupled to resin are available from Santa Cruz Biotechnology (#sc-8334AC).

Boric acid

Bromophenol blue

Chymostatin (Chemicon International, Temecula, CA, #E16)

Dimethylpimelimidate (DMP; Sigma-Aldrich, #D-8388)

DMSO

Dithiothreitol (DTT)

Ethanolamine

EDTA disodium salt (disodium EDTA•2H2O)

Ethylene glycol-bis(2-aminoethyl)-N,N,N′,N′-tetraacetic acid (EGTA)

Glycerol

Glycine

Hepes free acid

Leupeptin (Chemicon International, #E18)

Liquid nitrogen

Magnesium acetate

Magnesium chloride

Nocodazole

Nonidet P-40, 10% [NP-40, 10%; Roche, #1332473 (http://www.roche-applied-science.com)]

PBS, 10× (EMD Biosciences, #6505)

Pepstatin (Chemicon International, #EI10)

Polyoxyethylenesorbitan monolaurate (Tween 20)

Potassium chloride

Potassium hydroxide

Protease Inhibitor Cocktail Tablets, Complete, Mini, EDTA-Free (Roche, #1836170)

Sodium acetate

Sodium chloride

Sodium dodecyl sulfate (SDS)

Sodium hydroxide pellets

S protein agarose (Novagen, Madison, WI, #69704)

TEV protease [6×His-TEV protease(1 mg/ml stock), purified using nickel-nitrilotriacetic acid (Ni-NTA) agarose and gel filtration (11, 12), or available from Invitrogen [#12575-015 (http://www.invitrogen.com)]

Tris base

Urea (Invitrogen, #15505-035)

Plastics

0.5-ml and 1.5-ml microcentrifuge tubes

1.5-ml screw-cap tubes

15-ml and 50-ml polypropylene screw-cap tubes

Filter, Nalgene bottle-top, 0.2-μm pore size

TLA100.3 tubes [Beckman, #349622 (http://www.beckman.com)]

Equipment

Cell culture incubator, humidified, set at 37°C, 5% CO2

Cold room

Clinical centrifuge

Ultracentrifuge with TLA100.3 rotor (or equivalent)

Microcentrifuge

Sonicator with pulse capacity (e.g., Branson Digital Sonifier)

Rotator for microcentrifuge tubes

Mortar and pestle

Liquid nitrogen dewar flask

Metal wire strainer (flour sifter)

Recipes

Recipe 1: PBST
PBS, 10×50 ml
Tween-200.5 ml
Add ddH2O to these components to a final volume of 500 ml, mix, filter to sterilize, and store at room temperature.
Recipe 2: 1 M Sodium Borate, pH 9.0
Dissolve 61.8 g of boric acid in 800 ml of ddH2O and adjust the pH to 9.0 with NaOH pellets. Bring volume to 1000 ml with ddH2O and filter to sterilize. Store at room temperature. This is a 10× stock solution.
Recipe 3: 220 mM DMP
Allow DMP to come to room temperature for 20 min. Weigh out 34 mg of DMP and leave dry until just before use. Add 596 μl of 0.2 M sodium borate, pH 9 [diluted from 1 M Sodium Borate, pH 9.0 (Recipe 2)], mix to dissolve, and use immediately.
Recipe 4: 0.2 M Ethanolamine, 0.2 M NaCl, pH 8.5
Dissolve 12.2 g of ethanolamine and 11.7 g of NaCl in 800 ml of ddH2O, and adjust the pH to 8.5 with HCl. Bring volume to 1000 ml with ddH2O and filter to sterilize. Store at room temperature.
Recipe 5: HeLa Medium
DMEM440 ml
FBS50 ml
Penicillin-streptomycin, 100×5 ml
L-Glutamine, 100×5 ml
Combine components and filter to sterilize. Store at 4°C.
Recipe 6: 0.5 M EDTA
Dissolve 186.1 g of disodium EDTA•2H2O in 800 ml of double-distilled water (ddH2O) and adjust the pH to 8.0 with NaOH pellets (about 20 g). Bring volume to 1000 ml with ddH2O and sterilize by autoclaving. Store at room temperature.
Note: The EDTA will not go into solution until the pH approaches 8.0.
Recipe 7: PBS + 3 mM EDTA
Dulbecco’s PBS497 ml
0.5 M EDTA (Recipe 6)3 ml
Combine components and filter-sterilize. Store at room temperature.
Recipe 8: 1 M Hepes, pH 7.5
Dissolve 238.3 g Hepes free acid in 800 ml of ddH2O and adjust the pH to 7.5 with KOH. Bring volume to 1000 ml with ddH2O and sterilize by autoclaving. Store at room temperature.
Recipe 9: 0.5 M EGTA
Dissolve 190.2 g of EGTA in 800 ml of ddH2O and adjust the pH to 8.0 with KOH. Bring volume to 1000 ml with ddH2O and sterilize by autoclaving. Store at room temperature.
Recipe 10: 1 M MgCl2
Dissolve 203.3 g of MgCl2 in 800 ml of ddH2O. Bring volume to 1000 ml with ddH2O and sterilize by autoclaving. Store at room temperature.
Recipe 11: 2.5 M KCl
Dissolve 186.4 g of KCl in 800 ml of ddH2O. Bring volume to 1000 ml with ddH2O and sterilize by autoclaving. Store at room temperature.

Recipe 12: Freezing Buffer
Reagent Amount Final concentration
1 M Hepes, pH 7.5 (Recipe 8)25 ml50 mM
0.5 M EGTA (Recipe 9)1 ml1 mM
1 M MgCl2 (Recipe 10)0.5 ml1 mM
2.5 M KCl (Recipe 11)20 ml100 mM
Glycerol50 ml10%
Add ddH2O to a final volume of 500 ml, mix, filter to sterilize, and store at 4°C.
Recipe 13: Lysis Buffer
Reagent Amount Final concentration
1 M Hepes, pH 7.5 (Recipe 8)37.5 ml75 mM
0.5 M EGTA (Recipe 9) 1.5 ml1.5 mM
1 M MgCl2 (Recipe 10)0.75 ml1.5 mM
2.5 M KCl (Recipe 11)30 ml150 mM
Glycerol75 ml15%
10% NP-403.750.075%
Add ddH2O to a final volume of 500 ml, mix, filter to sterilize, and store at 4°C.
Just before use, to 5 ml add one Roche Mini EDTA-Free Complete Protease Inhibitor Cocktail Tablet.
Recipe 14: 0.1 M Glycine, pH 2.6
Dissolve 7.5 g of glycine in 800 ml of ddH2O and adjust the pH to 2.6 with HCl. Bring volume to 1000 ml with ddH2O and filter to sterilize. Store at 4°C.
Recipe 15: 1 M DTT
Dissolve 7.7 g of DTT in 50 ml of sterile ddH2O. Store in 1-ml aliquots at −20°C.
Recipe 16: 1000× LPC
Dissolve 100 mg of leupeptin, 100 mg of pepstatin, and 100 mg of chymostatin in 10 ml of DMSO. Store in 100-μl aliquots at −20°C.
Note: Alternative protease inhibitor cocktails, such as Complete EDTA-Free Protease Inhibitor Cocktail Tablets (Roche Applied Science, #1873580), can also be used.
Recipe 17: Wash Buffer
Reagent Amount Final concentration
1 M Hepes, pH 7.5 (Recipe 8)25 ml50 mM
0.5 M EGTA (Recipe 9)1 ml1 mM
1 M MgCl2 (Recipe 10)0.5 ml1 mM
2.5 M KCl (Recipe 11)60 ml300 mM
Glycerol50 ml10%
10% NP-402.5 ml0.05%
Add ddH2O to a final volume of 500 ml, mix, filter to sterilize, and store at 4°C.
Just before use, add 1 mM DTT [from 1 M DTT (Recipe 15)] and 1× LPC [from 1000× LPC (Recipe 16)].
Recipe 18: Cleavage Buffer
Reagent Amount Final concentration
1 M Hepes, pH 7.5 (Recipe 3)25 ml50 mM
0.5 M EGTA (Recipe 4)1 ml1 mM
1 M MgCl2 (Recipe 5)0.5 ml1 mM
2.5 M KCl (Recipe 6)60 ml300 mM
Glycerol50 ml10%
10% NP-402.5 ml0.05%
Add ddH2O to a final volume of 500 ml, mix, filter to sterilize, and store at 4°C.
Just before use, add 1 mM DTT [from 1 M DTT (Recipe 15)].
Recipe 19: 2 M Tris, pH 8.5
Dissolve 121.1 g of Tris base in 800 ml of ddH2O and adjust the pH to 8.5 with HCl. Bring volume to 1000 ml with ddH2O and sterilize by autoclaving. Store at room temperature.
Recipe 20: Pre-Urea Wash Buffer
Reagent Amount Final concentration
0.5 M EGTA (Recipe 9)1 ml1 mM
2.5 M KCl (Recipe 11)15 ml75 mM
2 M Tris, pH 8.5 (Recipe 19)12.5 ml50 mM
Add ddH2O to a final volume of 500 ml, mix, filter to sterilize, and store at room temperature.
Recipe 21: Urea Elution Buffer
Reagent Amount Final concentration
2 M Tris, pH 8.5 (Recipe 19)0.25 ml50 mM
Urea4.8 g8 M
Add sterile ddH2O to urea to bring to a volume of 9 ml. Vortex to mix. Add the Tris, and add ddH2O to 10 ml. Store at room temperature, but make fresh on the day of use.
Recipe 22: 1 M Tris, pH 6.8
Dissolve 121.1 g of Tris base in 800 ml of ddH2O and adjust the pH to 6.8 with HCl. Bring volume to 1000 ml with ddH2O and sterilize by autoclaving. Store at room temperature.
Recipe 23: Protein Sample Buffer (3×)
Reagent Amount Final concentration
SDS3 g6%
1 M Tris, pH 6.8 (Recipe 22)12 ml240 mM
Glycerol15 ml30%
Bromophenol blue20 mg ~0.04% (w/v) Final
Add sterile ddH2O to bring to a final volume of 50 ml. Store at room temperature, and add 50 μl of 100% 2-ME to 1 ml just before use.

Instructions

An overview of the purification strategy is shown in Fig. 1. After generation of a stable strain or cell line, sufficient material is prepared for the purification. Cells are lysed and insoluble material is removed by centrifugation. The fusion protein is then isolated using protein A Sepharose with antibodies against GFP. The GFP moiety of the fusion protein is then cleaved from the remaining portion using a site-specific protease, releasing the fusion into the supernatant. This protein is then isolated a second time using a second affinity resin (either S protein agarose or Ni-NTA). This combination of purification steps results in a highly pure sample that can then be analyzed by mass spectrometry.

We have developed two different LAP tag vectors to generate either N-terminal or C-terminal fusion proteins (Fig. 2 and Fig. 3). Although these vectors are similar in principle, the protease cleavage sites and the second affinity epitopes vary to provide a greater number of options for the purification strategy. For the purification protocol described below, we have focused on the use of the N-terminal GFP-TEV-S LAP tag. However, information for the purification of the C-terminal 6×His-PreScission-GFP LAP tag is included in the "Notes and Remarks" section. To generate a LAP tag for a protein of interest, the complete coding sequence should be inserted in-frame into the multiple cloning site.

Fig. 2.

Sequence of the pIC113 (pEGFP-TEV-S) tag. Green, pEGFP; blue, TEV cleavage site (19); light blue, S peptide (20). This vector is based on pEGFP-C1, which is kanamycin-resistant (Clontech). Sites listed (except XbaI) are unique to the vector.

Fig. 3.

Sequence of the pIC111 (6×His-PreScission-GFP) tag. Orange, 6×His; blue, PreScission cleavage site (2×) (21); green, EGFP; large arrow, transcription start site. This vector was constructed in pcDNA3.1, which is ampicillin-resistant (Invitrogen). Sites listed are unique to the vector.

It is essential to generate a stable cell line expressing the LAP fusion protein before conducting the purification. Stable cell lines can be constructed by various methods. The vectors pIC111 and pIC113 can be used directly by transfecting tissue culture cells and then selecting for neomycin resistance. Extensive descriptions of methods for generating stable cell lines with pEGFP-C1 and pcDNA3.1, upon which the LAP vectors are based, are described on the Clontech (http://www.bdbiosciences.com/clontech/techinfo/vectors/vectorsE/pEGFP-C1.shtml) and Invitrogen (http://www.invitrogen.com/content/sfs/manuals/pcdna3.1_man.pdf) Web sites, respectively. Alternatively, other methods, such as a retroviral-based vector (9, 13), can be used.

Following the generation of a stably transfected cell population as assessed by drug resistance, secondary screening should be conducted to obtain clonal cell lines with appropriate expression levels. The cells may first be sorted using either fluorescence-activated cell sorting (FACS) or examined by fluorescence microscopy to characterize GFP fluorescence. If FACS is used, clonal lines with different expression levels should be characterized microscopically after sorting to verify that the localization of the fusion protein is the same as that of the endogenous protein. In our experience, the highest expressers from the FACS profiles correspond to significant overexpression and improper localization of the fusion protein. By screening medium-to-low expressors from the FACS profile, it is possible to obtain four or five cell lines with sufficient amounts of the fusion protein and proper localization, as assessed by comparison to the endogenous protein. If the localization of the endogenous protein is unknown or diffuse, stable strains with different levels of expression can be compared in the purification procedure.

Coupling GFP Antibody to Protein A Beads

The instructions below apply to the preparation of one tube of antibody. Each LAP preparation will require four tubes. The volumes indicated for the beads are for the settled beads, not the slurry.

1. Wash 100 μl of Affi-Prep protein A beads three times with 1 ml of PBST (Recipe 1) in a microcentrifuge tube, centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s.

2. Resuspend the beads in 500 μl of PBST (Recipe 1) and add 55 μg of affinity-purified GFP antibody. Mix for 30 min to 1 hour by rotating at room temperature.

3. Wash beads three times with 1 ml of PBST (Recipe 1), centrifuging 10 to 15 s at full speed in a microcentrifuge between washes.

4. Wash beads three times with 1 ml of 0.2 M sodium borate, pH 9 [diluted from 1 M Sodium Borate, pH 9 (Recipe 2)], centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s.

5. Add 900 μl of 0.2 M sodium borate, pH 9 [diluted from 1 M Sodium Borate, pH 9 (Recipe 2)] to bring the final volume to about 1 ml.

6. Add 100 μl of freshly prepared 220 mM DMP (Recipe 3) and rotate tubes gently at room temperature for 30 min.

7. Wash beads two times with 1 ml of 0.2 M Ethanolamine, 0.2 M NaCl, pH 8.5 (Recipe 4), centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s.

Note: This treatment inactivates any residual cross-linker.

8. Resuspend in 1 ml of 0.2 M Ethanolamine, 0.2 M NaCl, pH 8.5 (Recipe 4) and rotate for 1 hour at room temperature.

9. Pellet the beads by centrifuging in a microcentrifuge at full speed for 10 to 15 s.

10. Resuspend beads in 500 μl of 0.2 M Ethanolamine, 0.2 M NaCl pH 8.5 (Recipe 4) and store at 4°C until use.

Note: Antibody-coupled beads are stable for several months at 4°C.

Preparation of Tissue Culture Cells for Biochemistry

1. Grow each clonal cell line expressing the LAP fusion protein in two 15-cm dishes with 20 ml of HeLa medium (Recipe 5) at 37°C, 5% CO2 until confluent.

2. Split each dish into 10 to 20 15-cm dishes, each with 18 ml of HeLa medium (Recipe 5).

Note: A total of 30 to 60 dishes yields approximately 3 to 5 g of cell material and will be required for one purification.

3. Grow at 37°C, 5% CO2 until about 90% confluent.

Note: To enrich for mitotic cells, add 100 ng/ml nocodazole to medium about 14 hours before harvesting cells. Prepare a 10× nocodazole stock solution in fresh medium, warm at 37°C for 5 min, and mix gently to make sure that it is in solution. Add 10× solution to existing media for a final concentration of 1×.

4. Remove the media from cells and pool the media in 50-ml conical tubes.

5. Add 3 to 4 ml of PBS + 3 mM EDTA (Recipe 7) to the cells and incubate for 5 min at 37°C.

6. Use the saved medium in the 50-ml conical tubes to wash the cells off the plates.

7. Transfer the cells in the medium and PBS to a 50-ml conical tube.

8. Pellet cells at 1000 rpm for 5 min at 4°C in a clinical centrifuge.

9. Wash cells with 50 ml of PBS, pelleting the cells at 1000 rpm for 5 min at 4°C in a clinical centrifuge.

10. Wash cells with 50 ml of Freezing Buffer (Recipe 12), pelleting the cells at 1000 rpm for 5 min at 4°C in a clinical centrifuge.

11. Resuspend cells in a small amount of Freezing Buffer (Recipe 12) and pipette the cells directly into liquid nitrogen in a clean dewar flask, allowing the cell suspension to form small beads as they enter the liquid nitrogen.

Note: The volume of Freezing Buffer should be just enough to resuspend the cells completely, about 2 ml per 4 ml of cell pellet.

12. Recover the cell suspension beads using a fine metal strainer (flour sifter) and store at −80°C until ready to grind.

13. Add the frozen cell suspension beads to the mortar and grind for about 1 min.

Note: Prechill the mortar and pestle for at least 5 min with liquid nitrogen before grinding.

14. Add additional liquid nitrogen and scrape the cells down to the middle of the mortar. Repeat two or three times until the cell material resembles a fine powder.

Note: Ground cell powder can be stored at −80°C indefinitely.

Preparation of Protein Extract

1. Weigh about 5 g of frozen ground cell powder, add an equal volume (about 5 ml) of Lysis Buffer (Recipe 13), and thaw in 37°C water bath until mostly melted.

Note: This Lysis Buffer is optimized for soluble cytoplasmic or nuclear proteins. It is important to adjust the final buffer for the protein of interest. For example, solubilization of membrane-associated proteins may require the addition of higher detergent concentrations.

2. Set up an ice-water bath and, with the sample immersed in it, sonicate the sample in a Branson Digital Sonifier using the following parameters:(1) 30% amplitude for 1 min in pulses(15 s on, 45 s off), then wait 2 min to allow sample to chill, and repeat entire sequence for a total of 3 min of sonication time; then(2) 40% amplitude for 30 s in pulses(15 s on; 45 s off).

Note: On different sonication machines, the sonication regime can be adjusted for optimal lysis as monitored using the Bradford colorimetric assay for protein concentration or UV absorption at 260 nm, which monitors nucleic acids released from cells. Full lysis is defined as the point where further sonication does not increase protein concentration or UV absorption.

3. Transfer crude extract to TLA100.3 tubes and spin at 20,000 rpm (16,500 g) for 10 min at 2°C, deceleration (DECEL) = 5.

4. Transfer supernatant to a fresh TLA100.3 tube and spin at 50,000 rpm (103,320 g) for 20 min at 2°C.

Note: Try to avoid collecting any lipids, and re-spin if this high-speed supernatant (HSS) is too cloudy.

5. Transfer the HSS to a 15-ml conical tube on ice and add KCl to a final concentration of 300 mM [for example, add 0.8 ml of 2.5 M KCl (Recipe 11) to 9.2 ml of HSS].

Note: The final amount of KCl present in the HSS and the wash steps below provides an important variable that can be adjusted for a specific purification. In general, we have found that 300 mM KCl has a dramatic effect on both the purity of the final product and the efficiency of the binding to the GFP antibodies, as compared to 100 mM KCl (under lower salt conditions, nonspecific binding to the protein A Sepharose-GFP antibody appears to compete with specific binding of the fusion protein). If a weak interaction is expected, lower salt conditions can be used. However, we have found the 300 mM salt conditions do not disrupt the vast majority of interactions.

GFP Immunoprecipitation

For a standard LAP preparation, use four tubes of protein A-coupled GFP antibody(400 μl of settled beads).

1. Wash each tube of coupled beads (about 100 μl) three times with 1 ml of 0.1 M Glycine pH 2.6 (Recipe 14), centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s.

Note: This step eliminates uncoupled or elutable antibodies and reduces background. Perform the washes quickly to avoid denaturing the antibodies.

2. Wash beads two times with 1 ml of Wash Buffer (Recipe 17) to neutralize the glycine, centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s.

3. Add 1 ml of Wash Buffer (Recipe 17) and divide each tube of beads into two separate tubes, for 8 tubes total.

4. Pellet the beads in a microcentrifuge at full speed for 10 to 15 s at 4°C and discard the wash buffer.

5. Mix each tube of beads with about 1 ml of HSS and rotate for 1 hour at 4°C.

6. Pellet the beads in a microcentrifuge at full speed for 10 to 15 s at 4°C, and transfer the HSS containing any unbound proteins to a fresh tube.

7. Drop-freeze the tube containing unbound HSS in liquid nitrogen, which can then be used for other purposes or repurification if binding is inefficient.

8. Rinse the beads three times with 1 ml of Wash Buffer (Recipe 17) at 4°C, centrifuging ~10 to 15 s at full speed in a microcentrifuge between washes.

9. Wash beads two times, incubating 5 min each wash with rotating at 4°C, in 1 ml Wash Buffer (Recipe 17), centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s.

10. Rinse beads two times with 1 ml of Cleavage Buffer (Recipe 18) at 4°C, centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s.

TEV Cleavage

1. Pool four tubes of HSS-immunoprecipitated GFP antibody beads into a single microcentrifuge tube (for a total of two tubes), and fill each tube with 1.3 ml of Cleavage Buffer (Recipe 18).

2. Add 30 μl of TEV protease(1 mg/ml stock) to each tube and rotate tubes overnight at 4°C.

Note: TEV is more active at 16°C than at 4°C. However, we prefer to use the lower temperature with a longer incubation time to prevent protein degradation.

3. Pellet the beads in a microcentrifuge at full speed for 10 to 15 s at 4°C and transfer the supernatants to fresh tubes.

4. Add 400 μl of Cleavage Buffer (Recipe 18) to beads to remove any residual protein, pellet in a microcentrifuge at full speed for 10 to 15 s at 4°C, and combine the supernatant with that from step 3.

Note: If the eluate contains residual protein A beads, recentrifuge and transfer to a new tube.

S Protein Agarose Purification

1. Wash two 80-μl tubes of S protein agarose slurry(40 μl of packed resin) three times with 1 ml of Wash Buffer (Recipe 17).

2. Add the TEV eluted supernatant to the S protein agarose and rotate for 3 hours at 4°C.

3. Pellet the beads in a microcentrifuge at full speed for 10 to 15 s at 4°C and discard the supernatant.

4. Wash beads three times with 1 ml of Wash Buffer (Recipe 17) at 4°C, centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s. Combine the resin into a single tube.

5. Wash beads two times with 1 ml of Freezing Buffer (Recipe 12) at 4°C, centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s.

Urea Elution

1. Wash beads two times with 1 ml of Pre-Urea Wash Buffer (Recipe 20) at room temperature, centrifuging between washes in a microcentrifuge at full speed for 10 to 15 s.

2. Transfer the beads to a 0.5-ml microcentrifuge tube using about 0.4 ml of Pre-Urea Wash Buffer (Recipe 20).

3. Pellet the beads in a microcentrifuge at full speed for 10 to 15 s and then carefully remove all residual supernatant using a fine gel-loading tip so as not to remove any resin.

4. Add 65 μl of Urea Elution Buffer (Recipe 21) to the beads and rotate for 30 min at room temperature.

5. Pellet the beads in a microcentrifuge at full speed for 10 to 15 s, remove 50 μl of urea to a fresh 0.5-ml tube, and drop-freeze this sample in liquid nitrogen.

Note: This urea-eluted sample provides an ideal starting point for analysis by mass spectrometry(4, 14, 15).

6. Add 10 μl of Protein Sample Buffer (Recipe 23) to remaining urea-eluted sample (10 to 15 μl) to check elution efficiency.

7. Add about 20 μl of Protein Sample Buffer (Recipe 23) to the remaining S protein agarose.

8. Boil samples for 3 min, run on an SDS-PAGE gel, and visualize proteins by silver staining.

Related Techniques

In addition to the purification strategy described here, numerous variations of the LAP tag exist that would be easy to implement and would increase the utility of this approach.

LAP Tagging in Other Organisms

The purification protocol described here was designed for use from samples prepared from cultured cells. However, the general LAP strategy should work in any organism in which it is possible to generate a stable transformant. Our lab has successfully utilized this strategy in C. elegans (9). For these purifications, the tagging construct, the generation of the stable C. elegans strain, and the growth of the worms were different from those described here for human cells. However, the purification protocol itself was identical to that used in cultured cells (9). It should be possible to adapt the tagging construct and the purification procedure to Escherichia coli, fungi, Arabidopsis, Drosophila, and most other organisms.

Use of 6×His-PreScission-GFP LAP Tag

For purification of the C-terminal construct (pIC111), PreScission protease (Amersham) should be substituted for the TEV protease and Ni-NTA resin (Qiagen) should be used instead of the S protein agarose. Bound proteins can be eluted with 250 mM imidazole or 100 mM EDTA and then lyophilized or TCA-precipitated for mass spectrometry. A similar strategy using a 6×His-PreScission-myc tag, including information on PreScission cleavage and elution from the Ni-NTA resin for mass spectrometry, has also been described (5).

Variations on the LAP Tagging Theme

In addition to the two LAP constructs described here, almost limitless combinations of tags exist that could be generated to take advantage of the unique features of other epitopes. For example, we have generated both yellow fluorescent protein (YFP)- and mRFP-based LAP tags, which open the possibility of two-color imaging of two LAP tags in a cell, or other fluorescent techniques, such as fluorescent resonance energy transfer (FRET). In each case, the primary reagent needed for the purifications is affinity-purified antibodies against the respective fluorescent protein. Because mRFP (16) represents a protein sequence that is distinct from the GFP variants [GFP, YFP, and cyan-fluorescent protein (CFP)], antibodies against mRFP and GFP do not cross-react, and thus the use of mRFP and GFP LAP tags would allow independent purification of two different LAP fusions protein coexpressed in the same cell.

Beyond Mass Spectrometry

The purpose of the purification protocol described above is to prepare samples that can be used to identify interacting polypeptides by mass spectrometry. However, there are a number of other uses for the purified protein.

The samples may be used to visualize protein profiles by gel electrophoresis or for Western blotting to test for interactions between the expressed LAP fusion and other proteins for which antibodies are available. In this case, Protein Sample Buffer (Recipe 23) can be added directly to the S protein agarose without performing the urea elution.

The samples may also be used to identify in vivo phosphorylation sites. In this case, a cocktail of phosphatase inhibitors(10 mM sodium pyrophosphate, 5 mM sodium azide, 10 mM sodium fluoride, 0.4 mM sodium orthovanadate, and 20 mM β-glycerophosphate) can be added to the initial Lysis Buffer (Recipe 13) to preserve any endogenous phosphorylations. Purified proteins eluted with urea can then be analyzed to identify posttranslational modifications (17).

Protein bound to the S protein agarose can be used for in vitro reactions such as kinase assays or ligand-binding assays. In these assays, the urea elution would not be performed, so that the proteins remained bound to the beads.

For assays in which soluble protein is required, the TEV eluate can be used before binding to S protein agarose. The TEV protease will not interfere with the majority of in vitro assays, and this eluate is typically pure enough for most experiments. If higher purity is desired, the TEV eluted protein can be further purified by gel filtration chromatography.

Notes and Remarks

GFP vs. LAP Tagging

The use of GFP to examine localization of a fusion protein in live cells is now common. Although it is possible that fusing the 26-kD GFP to the N or C terminus of a protein may interfere with some aspects of its function, genome-wide studies in budding yeast have demonstrated that more than 75% of the proteins are functional with a C-terminal GFP tag (8). The addition of the protease cleavage site and small second affinity tags used here for the LAP tag increase the total size of the fusion protein by only about 5 kD. Thus far, we have found both localization and overall level of fluorescence of GFP-tagged and LAP-tagged fusion proteins to be identical. Therefore, the LAP tag provides increased flexibility for analyzing protein function.

LAP vs. Other Purification Strategies

The purification protocol described here results in a final sample of very high purity. Typically, we find only one contaminating polypeptide that is present regardless of the nature of the fusion protein (usually an abundant heat shock protein). In our hands, the behavior of the LAP-based tandem affinity purification is virtually identical to standard TAP-tagging constructs [compare (9) to (18)]. In contrast, single-step immunoprecipitations have substantially higher background, presumably derived in part from nonspecific interactions with different polyclonal antibody populations.

Although the multistep purification procedure described here does result in high purity, especially as compared to single-step methods such as standard immunoprecipitations, it is possible that weakly interacting proteins may not persist through the more stringent LAP prep. In a comparison of LAP tagging and single-step immunoprecipitations with a set of known interacting proteins, we found one bona fide interacting protein that did not persist through the LAP purification (9). Varying the stringency of the LAP purifications by changing the salt concentrations during binding, although reducing purity, may help identify weaker interactions.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
View Abstract

Navigate This Article