Native Isolation of 3×HA-Tagged Protein Complexes to Characterize Protein-Protein Interactions

Abstract

Co-immunoprecipitation (Co-IP) is a straightforward method that is widely used in studying direct protein-protein interactions in physiological environments. This technique is based on the antigen-antibody interaction: the protein of interest (bait) is captured by a specific antibody, followed by antibody-bait precipitation. The proteins interacting with the bait protein (prey) co-precipitate with the antibody-bait complex from a cell lysate as an antibody-bait/prey complex. Nowadays, a variety of surface-functionalized materials with antibodies immobilized on agarose or magnetic beads are available, replacing the precipitation of antibodies and simplifying the application. However, unspecific binding of cellular proteins to matrix surfaces and/or antibodies has become a common issue. Unspecific binding that leads to false-positive signals and a high background can hamper further analysis. Our protocol describes a strategy to tremendously reduce unspecific background when isolating native proteins and protein complexes. Instead of eluting our samples under denaturing conditions, we elute triple hemagglutinin (3×HA)-tagged bait/prey complexes in their native form with a competitive peptide simulating the 3×HA tag of the bait protein. Matrix-unspecific interacting proteins and Co-IP antibodies remain on the matrix instead of being eluted under conventionally applied denaturing conditions. We optimized the elution by altering incubation time, eluent concentration, and temperature. These improvements result in more pure proteins. This strategy not only reduces background in SDS-PAGE and western blot but also allows complex characterization in vitro. © 2021 Wiley Periodicals LLC.

This article was corrected on 19 July 2022. See the end of the full text for details.

Basic Protocol:

The study of protein-protein interactions is one of the most important steps to understand the maintenance and regulation of biological activities in cells. There are numerous approaches to investigate protein-protein interactions in which co-immunoprecipitation (Co-IP) is used to analyze stable and strong protein complexes. For definition, the protein of interest is called bait protein. Proteins that interact with the bait protein and are co-precipitated are called prey proteins. Advantageous prey proteins can be unknown and are identified by Co-IP by analyzing the precipitates by mass spectrometry to identify novel interaction partners.

Formerly, protein-protein complexes were precipitated from cell lysates as large antibody-protein complexes. This strategy necessitates validated antibodies raised against each individual bait protein. Co-IP has been eased by fusing epitope tags to the bait protein at either the N- or the C-terminus, which facilitates the precipitation of a broad range of tagged bait proteins and their bound prey proteins with commercial high-affinity antibodies that bind to the fused epitope tag (Fig. 1). Another advantage of using tagged proteins is the fact that the epitopes that antibodies bind to are not always characterized. Antibody binding might compete with the bait/prey interaction, interfering with bait/prey binding by steric hindrance, if the antibody epitope is close to the bait/prey interface. Antibodies binding to the fused terminal epitopes are less likely to interfere with bait/prey binding.

Figure 1

Schematic overview of bead-assisted co-immunoprecipitation of 3×HA-E6AP/16E6/p53 ternary complex. The antibody captures the 3×HA-E6AP/16E6/p53 ternary complex by binding to the epitope fusion tag (yellow) of the 3×HA-E6AP. The antibody complex is then isolated by binding to a Protein G beaded support (left) or antibody directly immobilized on magnetic beads or agarose beaded support (right). Denatured complex proteins and antibodies (heavy and light chains) as well as proteins nonspecifically bound to the bead surface and antibodies (not shown) elute from the isolated beads under denaturing conditions. Protein G (left) and the heavy chain (right) are covalently coupled to the beads and do not co-elute. The red arrows indicate the possible site of native elution, which competes with the binding of the antibody to the epitope fusion tag. The pink arrow (left only) indicates the possibility to elute the complex by competing with the Protein G–antibody interaction. Here, the eluted antibody remains bound to the complex. Our strategy involves E6AP with N-terminal 3×HA fusion as an epitope for an anti-HA antibody, which is immobilized on magnetic beads. Native elution is triggered by a 3×HA-peptide competing with the antibody for fusion epitope binding, as indicated by the red arrow (right). The figure was created with BioRender.com.

Alternatively, antibody-protein complexes can be isolated by applying a Protein G or A affinity matrix. Proteins G and A have high affinity for the constant region (Fc) of most antibody classes, ensuring no interference with antigen binding to the variable site (Fab) of antibody molecules (Kaboord & Perr, 2008; Lin & Lai, 2017). Furthermore, the isolation of the antibody-bait/prey complex is simplified by direct, covalent coupling of antibodies to magnetic beads or agarose beaded support (Fig. 1). In combination with epitope tagging of the bait protein, this allows automation of the Co-IP. These alternative methods of isolating the antibody-bait/prey complex are not classical precipitations; rather, they are called pull downs.

Antibody interactions are very strong. Proteins can be eluted by a drop in pH or denaturing conditions such as in the presence of sodium dodecyl sulfate (SDS), urea, or guanidinium hydrochloride. Commonly, reducing SDS sample buffer is used, followed by SDS-PAGE and western blot analysis. All elution methods include the unfolding of proteins and disintegration of the protein complexes. Another disadvantage of denaturing elution is the co-elution of unspecific binders, immunoglobulin (Ig), or Ig light chains, which can cause problems in the downstream application, such as giving unspecific and false-positive signals or high background in western blots and mass spectrometry. Unspecific binding occurs due to binding of proteins to the antibodies or the surface of beads. Blocking the surface of the beads with bovine serum albumin (BSA) is described in many troubleshooting protocols to minimize unspecific binding of cellular proteins. However, this results in high BSA contamination after denaturing elution and interferes with mass spectrometry analysis. In addition, denaturing elution hinders further characterization of the eluted proteins, such as structural analysis, crosslinking/native mass spectrometry, or activity analysis. In order to tackle this problem, native elution of the protein complex, which does not elute unspecific binders, is necessary.

In this protocol, we describe a strategy for the elution of bait proteins under native conditions without loss of the proteins’ biological activities by employing the established 16E6/E6AP/p53 ternary complex as our experimental model. The method is based on the competitive elution of the triple hemagglutinin (3×HA)-tagged bait protein 3×HA-E6AP by applying 3×HA-peptide as an eluent. We analyzed eluted samples with respect to their protein content and purity by SDS-PAGE and western blot.

Overview and Principle

Co-IP is a direct method that is commonly used to study protein-protein interactions under physiological conditions. Antibodies bind to proteins of interest called bait proteins and facilitate co-purification of the interacting partners, called prey proteins, from cellular lysates (Fig. 1). Antibody-antigen interactions are one of the strongest reversible biological interactions. Most antibodies have a high avidity due to polyvalent epitope-binding sites. The dissociation constant (K_d) of an antibody-antigen interaction is usually in the picomolar (pM) range, demonstrating a high functional affinity. Furthermore, antibodies are highly specific regarding antigen binding. Consequently, antibodies can bind to their antigens even at low antigen concentrations in crude extracts, making them preferable candidates for affinity purification. However, for a successful Co-IP, it is important to maintain the integrity of proteins and protein complexes during the Co-IP procedure. Thus, gentle lysis and appropriate washing conditions are required (see Strategic Planning).

The antibody-bait/prey complex can be isolated by different approaches, including direct precipitation, binding to Protein G or A, or direct covalent coupling of the bait-specific antibody to a beaded support (Fig. 1). Proteins G and A are bacterial cell membrane proteins that bind to Ig heavy chains, preferring IgG and IgA, respectively. Commonly, Proteins G and A are covalently coupled to a beaded support. Usually, this strategy is used if the bait is not fused to an epitope tag. Consequently, the Co-IP needs to involve a bait protein–specific antibody.

Magnetic beads with immobilized antibodies are commonly used for Co-IP of proteins with epitope fusions, such as HA, FLAG, c-myc, or V5 tags.

Affinity Epitope Tagging

Fusing epitope tags to a bait protein is commonly used in detecting proteins in vitro and in cell culture. These epitopes are relatively short and have very specific primary antibodies that are commercially available for most tags. These short tags also rarely affect the properties of the heterologous protein of interest. HA peptide (YPYDVPDYA) derived from human influenza virus HA protein is one of the fusion protein tags widely used in recombinant protein expression and immunodetection in mammalian cells. We choose the multimeric HA tag (3×HA) because it was previously shown by Ranawakage, Takada, & Kamachi (2019) that this trimeric tag can significantly improve the affinity toward an anti-HA antibody, specifically six-fold. In the case of bait proteins that show low expression levels, this trimeric tag can be a key means of improving Co-IP. Commercially, several anti-HA antibodies that are pre-immobilized on magnetic beads or agarose resin beads are available, allowing isolation of HA-tagged protein complexes from cell lysate.

Native Elution Improves Specificity and Purity of Protein Complexes

Cell lysates are a complex mixture of macromolecules such as nucleic acids, proteins, and lipids. These macromolecules can bind nonspecifically to the Fc region of an antibody or to the surface of functionalized beads. Importantly, antibodies are highly specific, but cross-reactivity might occur toward proteins with similar epitopes. Unspecific binding can result in false-positive signals and increase the background and therefore hamper further analysis dramatically. Denaturing elution of protein complexes, commonly with reducing SDS sample buffer, not only elutes the bait/prey complex but also elutes nonspecifically bound macromolecules and antibodies. To overcome these common issues in Co-IP, it is more desirable to recover proteins or protein complexes in their native form. Previously, several peptide reagents have been reported to be successful in natively eluting protein complexes containing Protein A–tagged baits. These peptides are able to competitively bind to the hinge region on the Fc domain of IgG, thus releasing Protein A–tagged bait/prey complexes (Strambio-de-Castillia, Tetenbaum-Novatt, Imai, Chait, & Rout, 2005; LaCava, Chandramouli, Jiang, & Rout, 2013; LaCava, Fernandez-Martinez, Hakhverdyan, & Rout, 2016). Applying these peptides for native elution after Co-IP as illustrated in Figure 1, the complex would theoretically remain in its native form, but the antibody targeting the bait would still be bound to the eluted complex (see Fig. 1).

In this article, we describe Co-IP of the known ternary complex E6/E6AP/p53. Ubiquitin-protein ligase E3A, also known as E6AP, is N-terminally fused to a 3×HA-peptide (H-YPYDVPDYA YPYDVPDYA YPYDVPDYA-OH) as a bait. After Co-IP, we elute 3×HA-E6AP by competing with the binding of anti-HA monoclonal antibodies to 3×HA-tagged baits using a synthetic 3×HA-peptide, and we expect E6 and p53 to co-elute. In order to optimize this native elution, we consider several factors influencing the yield of elution: (i) Temperature: Temperature affects protein stability. In general, the higher the temperature, the more unstable the protein is. Even though mammalian cell culture is performed at 37°C, isolated proteins are not in their natural environment after cell lysis. Researchers always try to mimic physiological conditions; however, the stability of isolated proteins is usually lower. In vitro, experimental temperatures should always be evaluated with care to avoid protein aggregation and artifacts. In addition, the temperature has a kinetic effect on chemical reactions. According to the Arrhenius equation and van't Hoff's law, a general rule of thumb is that if the temperature is increased by 10 Kelvin (K), the reaction is 2 to 3 times faster. In summary, the temperature influences protein stability and elution time. (ii) Elution time: The time that the competition needs is dependent on the rate of dissociation of the antibody-3×HA-bait complex and the rate of formation of the antibody-3×HA-peptide complex. Both rates depend on the temperature, as described above. It is desirable to find a balance between protein stability and fast and complete reaction to optimize the yield of native elution. (iii) Eluent concentration: The 3×HA-tagged bait/prey complexes are eluted from the anti-HA antibody competitively by addition of 3×HA-peptide. The affinity of the anti-HA antibody for 3×HA-epitope is very high, with a K_d in the pM range. The excess of the 3×HA-peptide must be high enough for it to compete with the 3×HA-bait and depends on the antibody concentration, the 3×HA-bait concentration, and buffer conditions. (iv) Co-factors: To isolate native proteins and complexes, it is necessary to employ all co-factors that the proteins and complexes need. Otherwise, the complexes may dissociate, and the proteins may aggregate.

Strategic Planning

In general, it is recommended to perform a conventional Co-IP with denaturing elution to verify the proof of principle and to verify the parameters for strategic planning before starting a native elution experiment.

Selection of an epitope tag

There are two factors that one should consider in selecting an epitope tag for the bait protein: cross-reactivity of the tag with cellular proteins and whether or not the tag would hinder biological activity of the bait. If the N-terminus of a bait participates in prey binding, one may consider fusing the epitope at the C-terminus and vice versa.

Selection of bait

If one wants to purify a native complex of proteins that one already knows to interact, which component of the complex is chosen as the bait might be critical. In general, one should consider the expression level of the 3×HA-bait, the expression levels of endogenous preys, and protein stability.

To pull down the ternary complex E6/E6AP/p53, we choose E6AP as bait because the endogenous expression of E6AP is too low for detection and heterologous expression is necessary. The endogenous expression level of p53 is already high enough. 16E6, as a viral protein, also needs to be expressed heterologously. However, we did not choose to use 16E6 as the bait because both p53 and E6AP proteins directly bind to 16E6, and we intended to show whether native elution is suitable to isolate complexes even with indirect interaction partners, as is supposed for E6AP and p53. It is very important to perform test expression in intended experimental cell lines to ensure good expression of the 3×HA-bait protein and prey proteins (if known) for an efficient pull down.

If protein stability is an issue due to proteasomal degradation, proteasome inhibitor can be applied in cell culture. We need the addition of proteasome inhibitor (MG132) because we know that HPV16E6 recruits E6AP, forming a ternary complex with p53 that then leads to the proteasomal degradation of p53. Addition of MG132 can prevent the degradation of p53.

Buffer design

Components of lysis buffers, washing buffers, and elution buffers should always be questioned with respect to protein complex demands. For example, ethylenediaminetetraacetic acid (EDTA), as a chelator of bivalent cations, is commonly used to block metalloproteases during cell lysis. When a protein complex requires bivalent cations, EDTA should be avoided. For isolation of the ternary complex E6/E6AP/p53, we know that E6 and p53 are zinc-finger proteins, i.e., they bind zinc ions via cysteines of their thiol groups. As a consequence, we do not use EDTA in any buffer, and we always supplement buffers with 1 mM Tris(2-carboxyethyl)phosphine (TCEP) as a reducing agent. Dithiothreitol (DTT) and β-mercaptoethanol (b-ME) could also be used as reduction agents. They are less bulky but also less stable than TCEP. Generally, unspecific oxidation of cysteine-rich proteins should be avoided by adding reducing agents because this can lead to false-positive interaction signals. Notably, if disulfide formations are crucial for protein structure and stability, reducing agents should be omitted. Additionally, Benzonase^® should always be included in lysis buffer, as it could reduce viscosity caused by nucleic acids and prevents cell clumping by rapidly hydrolyzing all forms of DNA and RNA. An appropriate amount of bivalent magnesium ions must be included to allow an optimal level of Benzonase^® activity.

Cell lysis

Cell lysis is a critical step. All cellular compartments are mixed, resulting in accessibility to phosphatases, proteases, and pH shifts. Phosphorylation can be important for protein interactions. Proteases can degrade the proteins of interest, and pH shifts can result in loss of function. Consequently, lysis buffers should be supplemented with a buffer reagent maintaining the demanded pH and containing protease inhibitors and phosphatase inhibitors. Additionally, it is important to consider cell lysis agents. Detergents, depending on the type and concentration, might interfere with the protein structure, leading to aggregation or denaturation, but may solubilize the protein. This can lead to low efficiency of the pull down due to a loss of function of bait or prey proteins. We carry out all lysis steps on ice to avoid any protein aggregation because this might lead to co-aggregation of the proteins of interest.

Necessary controls

Bait and prey proteins can also nonspecifically interact with matrix surfaces and antibodies. Though unspecific interactions are minimized by applying native elution, it is recommended to include the following controls, depending on if the prey is an endogenous protein or overexpressed:

Bait-only control: cells transfected with bait only (3×HA-E6AP) → Co-IP signal for bait only (3×HA-E6AP, no signal for p53 and E6)

Overexpressed-prey control: cells transfected with prey only (E6) → no Co-IP signal for bait and prey

Endogenous-prey control: cells transfected with empty plasmid (mock) → no Co-IP of endogenous proteins

Please note that we do not include the mock to check unspecific binding of p53 because p53 does not bind to E6AP; it binds to E6 if complexed with E6AP. Consequently, a negative Co-IP signal for p53 in the bait-only control is adequate to check for unspecific binding of p53.

Eluent

In analogy to the 3×HA fusion epitope, a 3×HA-peptide is used. The 3×HA-peptide (H-YPYDVPDYA YPYDVPDYA YPYDVPDYA-OH) is a synthetic peptide of 27 amino acid residues used to competitively elute HA-tagged fusion proteins that bind to anti-HA antibodies conjugated on magnetic beads or agarose resin beads. It is necessary to synthesize this peptide with purity of >90% in order to obtain a specific elution of target native protein complexes with high yields.

Elution conditions

It is very important to optimize the elution procedure for maximal efficiency. With respect to the described elution parameters, the temperature, elution time, and eluent concentration were set as variables for optimization. To screen these various conditions, we always loaded the magnetic beads with the same amount of cell lysate according to the total protein content. This allowed a comparable load for each experimental setup assuming similar expression of the proteins. The total protein content was determined by a Bradford measurement in triplicate for each sample.

For temperature tests, it is important that all steps (incubation and elution) are carried out at the respective temperature. Otherwise, the kinetic experiment will be misinterpreted and be less reproducible.

Native Isolation of 3×HA-Bait Protein Complex

Cells co-transfected with 3×HA-E6AP and HPV16E6 are lysed, followed by affinity capture with anti-HA microbeads using a μMACS HA Isolation Kit. Bound proteins are loaded onto μColumns that are placed in the magnetic field of a μMACS Separator. Unbound proteins are washed away before 3×HA-peptide is applied for native elution. Denaturing elution is applied to elute residual proteins that are retained on the anti-HA microbeads to determine elution efficiency for further optimization processes. Figure 2 summarizes the workflow of this protocol.

Figure 2

Schematic diagram illustrating the steps of native isolation of 3×HA-bait protein complex. Lyse transfected cells that express 3×HA-bait protein. Isolate 3×HA-bait protein (yellow-blue) and its interacting proteins (prey in red) with magnetic beads (purple) conjugated with anti-HA antibody (dark blue) and wash away unbound protein in a magnetic field. Elute 3×HA-bait/prey complex from the antibody-conjugated beads by applying 3×HA-peptide (green). The 3×HA-peptide concentration, elution temperature, and elution time need to be optimized (steps 4 to 7). Eluted complexes are analyzed by SDS-PAGE, western blot, or mass spectrometry or other biophysical methods (not shown) to analyze the structure and function of the complex. The figure was created with BioRender.com.

Materials

• HEK293T cells (for our work, kind gift from Dr. Murielle Masson, IGMBC)
• Dulbecco's modified Eagle's medium (DMEM; Gibco^® by Life Technologies, #41965-062) supplemented with 10% (v/v) fetal bovine serum (FBS; Gibco^® by Life Technologies, #10270-106) and 50 μg/ml gentamicin (Gibco^® by Life Technologies, #157710049), 37°C
• 3×HA-E6AP plasmid DNA (GenScript)
• HPV16E6 pcDNA3.1 (+) plasmid DNA (GenScript)
• Polyethyleneimine 25K (PEI 25K; Polysciences, #23966-1)
• Opti-MEM™ Reduced Serum Medium with GlutaMAX Supplement (Opti-MEM™; Gibco^® by Life Technologies, #51985034)
• 10 mM MG132 proteasome inhibitor (AdipoGen Life Sciences, #AG-CP3-0011)
• 3×HA-peptide (Intavis Peptide Services)
• Peptide buffer (see recipe)
• 10 M NaOH
• Liquid nitrogen
• Dulbecco's phosphate-buffered saline (DPBS; Gibco^® by Life Technologies, #14190-169), 4°C
• Lysis buffer (see recipe), 4°C
• Bradford reagent (see recipe)
• Anti-HA microbeads (from μMACS HA Isolation Kit, Miltenyi Biotec, #130-091-122), 4°C
• 1× reducing SDS sample buffer (from 5× reducing SDS sample stock buffer; see recipe), 95°C
• 8% to 20% SDS-PAGE gel
• 1× SDS running buffer (from 5× SDS running stock buffer; see recipe)
• Coomassie staining solution (see recipe)
• De-staining solution (see recipe)
• Western blot transfer buffer (make fresh; see recipe)
• 5% (w/v) bovine serum albumin (BSA; Serva, #9048-46-8) in 1× phosphate-buffered saline (PBS; from 10× PBS; see recipe)
• 0.05% (v/v) and 0.1% (v/v) Tween 20 in 1× PBS (from 10× PBS; see recipe) (PBS-T)
• Primary antibodies:
  • Anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibody (6C5; mouse; Santa Cruz Biotechnology, #sc32233)
  • Anti-HA monoclonal antibody (C29F4; rabbit; Cell Signaling, #3724)
  • Anti-HPV16E6 antibody (mouse; for our work, kind gift from Arbor Vita Corporation)
  • Anti-p53 antibody (DO-1; mouse; BioLegend, #645702)
• Secondary antibodies:
  • IRDye^® 680RD goat anti-mouse antibody (LI-COR Biotechnology GmbH, #926-68070)
  • IRDye^® 680RD goat anti-rabbit antibody (LI-COR Biotechnology GmbH, #926-68071)

• 100- and 150-mm sterile cell culture dishes (Thermo Scientific)
• 37°C, 95% humidity, and 5% CO₂ incubator
• Vortex mixer
• NanoDrop spectrophotometer
• Eppendorf microtubes
• 15-ml conical centrifuge tubes (Falcon™), 4°C
• Refrigerated centrifuge, 4°C
• Shaker
• μColumns (Miltenyi Biotec, #130-042-701)
• μMACS Separator (Miltenyi Biotec, #130-042-602)
• Microcentrifuge
• Nitrocellulose membranes (GE Healthcare, #10600001)
• LI-COR Odyssey Fc or equivalent imaging system
• ImageJ 1.47v

• Additional reagents and equipment for Bradford assay (see Current Protocols article; Simonian & Smith, 2006), SDS-PAGE and Coomassie blue staining (see Current Protocols article; Gallagher & Sasse, 2001), and western blotting (see Current Protocols article; Gallagher, Winston, Fuller, & Hurrell, 2008)

NOTE: All solutions and equipment coming into contact with cells must be sterile, and proper sterile technique should be used accordingly.

Transfection of plasmid DNA expressing 3×HA-bait protein

1. Culture HEK293T cells in 10 ml DMEM supplemented with 10% FBS and 50 μg/ml gentamicin in 100-mm sterile cell culture dishes at 37°C, 95% humidity, and 5% CO₂. One day before transfection, seed 8 × 10⁶ HEK293T cells in 25 ml medium in 150-mm sterile cell culture dishes, with four dishes for each sample.

2. Replace medium with 25 ml fresh medium (same as in step 1) before transfection starts. Then, transfect cells with 12 μg of 3×HA-E6AP plasmid DNA and 8 μg HPV16E6 pcDNA3.1 (+) plasmid DNA using PEI 25K in Opti-MEMn™ at a DNA/PEI 25K ratio of 1:3 following the manufacturer's instructions.

3. Treat cells with 3 μM MG132 proteasome inhibitor (from 10 mM stock) 8 hr post-transfection.

Preparation of 3×HA-peptide

4. Prepare a 5 mM stock concentration of 3×HA-peptide by dissolving peptide in peptide buffer and adjust pH to 7.0 with 10 M NaOH.

Steps 4 to 7 should be carried out 1 day before use in step 17 or during the incubation time in step 13.

5. Vortex vigorously to ensure that the peptide is completely dissolved and measure final concentration of peptide at an absorbance of 280 nm using a NanoDrop spectrophotometer, with peptide buffer as a blank.

6. Calculate concentration of the dissolved peptide based on the Beer's Law equation, A₂₈₀ = ε₂₈₀lc, where

• A₂₈₀ = absorbance (AU);
• Ε₂₈₀ = molar extinction coefficient, in this case 13 410 M^-1cm^-1 (retrieved from ExPASy ProtParam tool);
• l = length of the path that light must travel in the solution in cm (usually 1 cm); and
• c = concentration of the peptide solution [molar (M)].

7. Aliquot peptide stock solution in Eppendorf microtubes, freeze in liquid nitrogen, and store at −80°C for long-term storage to avoid hydrolysis of the peptide.

Cell lysis

8. Twenty-four hours post-transfection, harvest transfected cells by flushing the cells directly with 3 ml cold DPBS per dish, transferring cells to a pre-cooled 15-ml conical centrifuge tube, and centrifuging 10 min at 500 × g, 4°C.

IMPORTANT NOTE: Starting here, all steps should be carried out on ice or in a cold room.

9. Discard supernatant and wash cell pellet once again in 10 ml cold DPBS followed by centrifugation for 10 min at 500 × g, 4°C.

10. Resuspend cell pellets in 3 ml cold lysis buffer and incubate cell lysate on a shaker for 1 hr in a cold room.

11. Centrifuge cell lysate for 10 min at 16,000 × g, 4°C, to remove cell debris.

12. Determine total protein concentration of the cell lysate using Bradford reagent (see Current Protocols article; Simonian & Smith, 2006). Save an aliquot of cell lysate for analysis by western blot (input; see step 24).

Isolation of 3×HA-bait/prey complexes by antibody binding

13. Add 50 μl cold anti-HA microbeads to cell lysate and incubate on shaker for 4 hr in the cold room.

14. After incubation, place a μColumn in magnetic field of a μMACS Separator and equilibrate column with 1 ml cold lysis buffer.

15. Apply cell lysate onto the μColumn and allow to run through.

Flow-through could be collected for further analysis by western blot. This step is recommended if the pull-down efficiency is not known or has low reproducibility.

16. Rinse column three times, each time with 500 μl cold lysis buffer. Proceed immediately to step 17.

Native elution by competitive displacement with 3×HA-peptide

17. Prepare 3×HA-peptide (see step 7) at appropriate working concentration (see Understanding Results section) and then apply 120 μl to the μColumn and immediately remove column from the μMACS Separator.

18. Collect flow-through in Eppendorf microtubes.

19. Reapply peptide solution onto μColumn 10 times before incubating solution on a shaker at 300 rpm in the cold room (see Understanding Results section for incubation time). Keep μColumn for use in step 20.

20. After incubation, equilibrate μColumn with 500 μl cold peptide buffer on the μMACS Separator. Then, apply peptide solution onto the μColumn, 60 μl at a time, and allow it to run through into a new Eppendorf microtube. Keep an aliquot for western blotting analysis.

21. Add 60 μl pre-heated 1× reducing SDS sample buffer onto μColumn and immediately remove column from the magnetic field to elute residual proteins retained on the μColumn.

22. Incubate μColumn at room temperature for 5 min before adding another 60 μl pre-heated 1× reducing SDS sample buffer.

23. Spin down μColumn in a new Eppendorf microtube for 20 s at 250 × g at room temperature in a microcentrifuge (denaturing elution).

SDS-PAGE and western blot

24. Resolve all protein samples on a 8% to 20% SDS-PAGE gel with 1× SDS running buffer and stain gel with Coomassie staining solution for 30 min on a shaker at room temperature followed by de-staining solution (see Current Protocols article; Gallagher & Sasse, 2001) or transfer protein onto a nitrocellulose membrane using for 1 hr at 70 V or 90 min at 90 V depending on the Western blot transfer buffer used (see recipe) (see Current Protocols article; Gallagher et al., 2008). For de-staining, change de-staining solution every hour, until blue bands and a clear background are obtained.

25. Block membrane with 5% BSA in 1× PBS on a shaker for 1 hr at room temperature.

26. Incubate membrane with primary antibodies in 0.1% PBS-T overnight on a shaker in the cold room (or according to the manufacturer's recommendation) and then wash three times with 0.05% PBS-T before applying secondary antibodies in 0.1% PBS-T for 30 min on a shaker at room temperature.

Antibodies should be diluted according to the manufacturer's instructions. We dilute all primary and secondary antibodies in this protocol with 0.1% PBS-T.

27. Wash membrane three times with 0.05% PBS-T and visualize for respective signals on a LI-COR Odyssey Fc or equivalent imaging system.

28. Analyze densitometry of the protein signals using ImageJ 1.47v. Calculate elution efficiency using the equation given below:

$$\begin{eqnarray*} &&Native\ elution\ \it e\!f\!f\!iciency\ \left( \% \right)\\ &&= \frac{{{\rm{signal\ of\ native\ elution}}}}{{{\rm{signal\ of\ native\ elution\ }} + {\rm{\ signal\ of\ denaturing\ elution}}}}\ \times \ 100\% \end{eqnarray*}$$

REAGENTS AND SOLUTIONS

Bradford reagent

Dissolve Coomassie Brilliant Blue G-250 in 5% (v/v) methanol to 0.05 g/L. Slowly add 85% (v/v) phosphoric acid to this solution mix to 10%. Then, slowly add this acid solution mix to 500 ml water and mix with a magnetic stirrer. Add water to 1 L. Store ≤1 year at 4°C in the dark.

IMPORTANT NOTE: Do NOT add water to the acid solution. The solution should be filtered through Whatman paper (GE Healthcare, #10426890) to remove any precipitate.

CAPS buffer

• 10 mM 3-(cyclohexylamino)-1-propanesulfonic acid (CAPS)
• 0.001% (w/v) SDS
• 10% (v/v) methanol
• Adjust to pH 10.3

Coomassie staining solution

• 10% (v/v) acetic acid
• 0.05% (w/v) Coomassie Brilliant Blue R-250
• 25% (v/v) propan-2-ol

• Store ≤6 months at room temperature

  This is an in-house Coomassie staining solution commonly used in our lab. Commercially available staining solution could also be used following the manufacturer's instructions.

De-staining solution

• 10% (v/v) acetic acid
• 10% (v/v) ethanol
• 80% (v/v) deionized water

• Store ≤6 months at room temperature

  This is an in-house de-staining solution commonly used in our lab after Coomassie staining. Commercially available de-staining solution could also be used following the manufacturer's instructions.

Lysis buffer

• 10% (v/v) glycerol
• 50 mM HEPES, pH 7.5
• 3 mM magnesium chloride (MgCl₂)
• 0.1% (w/v) Nonidet P-40 or IGEPAL CA-630 (NP-40)
• 150 mM sodium chloride (NaCl)
• 1 mM TCEP
• 200 μM zinc chloride (ZnCl₂)
• Store ≤1 month at 4°C

• Supplement with Benzonase^®, phosphatase inhibitor, and protease inhibitors immediately prior to cell lysis

  This buffer recipe was proven to be useful for zinc-finger proteins in our lab, but it is not universally useful and may require modification, such as use of a different buffer agent, pH, redox state, or detergent according to the protein of interest. Substitution of HEPES with Tris should be avoided if mass spectrometry is involved in the downstream process. Besides, one must be aware that the pH of Tris-based buffer is temperature dependent. We recommend the PhosSTOP™ tablet and cOmplete™ EDTA-free Protease Inhibitor Cocktail from Roche and Benzonase^® endonuclease with purity grade II (>90%) from Merck. Use of EDTA-free protease inhibitors is crucial for protein binding bivalent cations as the presence of EDTA, even in limited amount, would lead to chelation of bivalent ions present in the solution.

Peptide buffer

• 10% (v/v) glycerol
• 50 mM HEPES, pH 7.5
• 150 mM NaCl
• Store ≤1 month at 4°C

Phosphate-buffered saline (PBS), 10×

• 2 g/L potassium chloride (KCl)
• 2 g/L potassium dihydrogen phosphate (KH₂PO₄)
• 80 g/L NaCl
• 11.5 g/L sodium dibasic phosphate (Na₂HPO₄)
• Adjust pH to pH 7.2 with 85% (v/v) phosphoric acid
• Store ≤6 months at room temperature

Reducing SDS sample stock buffer, 5×

• 5% (v/v) b-ME
• 0.005% (w/v) bromophenol blue
• 50% (w/v) glycerol
• 5% (w/v) SDS
• 0.25 M Tris·HCl, pH 8
• Make 950-μl aliquots from 50 ml stock buffer
• Store aliquots ≤6 months at −20°C

• Add 50 μl b-ME to aliquot immediately before use

  This should be used at 1× concentration in the final sample. Commercially available reducing SDS sample buffer could also be used following the manufacturer's instructions.

SDS running stock buffer, 5×

• 960 mM glycine
• 1% (w/v) SDS
• 125 mM Tris·HCl, pH 8.3

• Store ≤6 months at room temperature

  This should be used at 1× concentration.

Western blot transfer buffer

• Option 1:
• 1.2 g/L CAPS buffer (see recipe)
• 10% (v/v) methanol
• 1 g/L SDS
• Adjust pH to 10.3 with 10 M NaOH
• Prepare fresh immediately before use
• Option 2:
• 11.26 g/L glycine
• 10% (v/v) methanol
• 0.2 g/L SDS
• 2.44 g/L Tris·HCl, pH 8.2

• Prepare fresh immediately before use

  There are several different formulations of western blot transfer buffer available. Above, we provide two formulations that are commonly used in our lab. For option 1, run the transfer at 70 V for 60 min, as optimized for the detected proteins. For option 2, run the transfer at 90 V for 90 min, as optimized for the detected proteins. In both cases, these parameters might be changed with respect to the protein of interest.

COMMENTARY

Background Information

Please see the Overview and Principle section above for background information.

Critical Parameters

First of all, make sure that the 3×HA-peptide has purity of >90%, and the concentration of the dissolved peptide should always be verified spectroscopically because this value directly influences the yield of native protein elution.

The incubation and elution temperatures and times should always be optimized based on the stability of the proteins of interest. Appropriate buffer conditions, especially pH and use of an appropriate detergent for cell lysis or required co-factors, ensure the stability and functionality of the proteins of interest.

Troubleshooting

Please refer to Table 1 for a troubleshooting guide for natively eluting protein complexes.

Table 1. Troubleshooting Guide for Native Isolation of Protein Complexes

Problem	Possible cause	Solution
Poor peptide solubility	Peptide solution is too acidic	Adjust pH of peptide carefully with 10 M sodium hydroxide and dissolve at neutral pH (∼7)
Low elution efficiency: first perform Co-IP under denaturing conditions for proof of principle: Positive result: problem caused by native elution → continue troubleshooting Negative result: problem is principally Co-IP problem: refer to Current Protocols article Brizzard & Chubet (1997)	Proteins are unstable due to improper buffer components	Refer to Strategic Planning (Buffer design and Cell lysis) and annotation to recipe for lysis buffer
	Proteins are temperature sensitive	Carry out all steps at low temperature (4-8°C)
	Proteins are unstable over time	Optimize incubation time for native elution
	Poor peptide quality	Use only peptide with purity >90%
	Insufficient amount of peptide used	Titrate peptide carefully and work with saturated concentration. Re-check peptide concentration spectroscopically.
	Peptide was hydrolyzed due improper storage and handling	Freeze peptide solution in liquid nitrogen and store at −80°C. Store lyophilized peptide at −20°C. Aliquot peptide solution to avoid repeated freeze-thaw cycles.
Low signal of prey	Low amount of prey in cell lysate	Overexpress prey for complex isolation

Please note that Co-IP including the presented native elution requires multiple steps and that variations can occur, starting from plasmid preparation, transfection, and peptide solubilization down to protein transfer to nitrocellulose membranes and protein probing. Variations influence the input, elution, and protein detection. Hence, Co-IP is not a quantitative method. As a consequence, semiquantitative comparisons between different experiments are questionable. From our experience, the most reproducible results can be obtained by splitting cell extracts for elution tests and running all samples of interest on one SDS-PAGE gel and blot membrane. Freezing of cell lysates and eluents affects elution efficiency and western blot analysis, respectively.

Understanding Results

3×HA-peptide stock solution

Peptide was dissolved to get a theoretical 5 mM stock concentration, as described above, in peptide buffer. A final concentration of 4.71 mM, which is equivalent to 94.2% of solubility, was obtained. This is a reasonable yield, and it indicates high solubility of the applied peptide under the described conditions. However, one should always verify the peptide concentration spectroscopically because the yield of dissolved peptide directly influences the yield of native protein elution.

Optimizing elution conditions

As described above, the native elution (steps 17 to 19, Fig. 2) was optimized with regard to optimal temperature, elution time, and eluent concentration. For optimization, the analysis was based on the bait protein 3×HA-E6AP only, assuming that bound prey protein does not interfere with the competitive elution of the epitope fusion tag. As a reminder, first, we performed a native elution and then applied the same beads for denaturing elution in order to analyze the bait protein that remains on the beads. The elution efficiency was analyzed by (a) performing a western blot of the samples eluted under the different conditions, (b) detecting the eluted protein 3×HA-E6AP using a fluorescently labeled anti-HA-antibody and a fluorescence imaging system, and (c) quantifying signal bands densitometrically (Fig. 3).

Figure 3

Effect of temperature on native elution of 3×HA-E6AP. (A) Cell lysates of HEK293T cells overexpressing 3×HA-E6AP were divided equally, corresponding to 2500 μg/ml total protein. Each was incubated with 50 μl magnetic anti-HA microbeads. The bead solution was transferred to a μColumn, a magnetic field was applied, and unbound proteins were washed away as described above (Fig. 2). 3×HA-E6AP complex was eluted (step 9, Fig. 2) with 120 μl of 1200 μM 3×HA-peptide at 4°C, 25°C, or 37°C for 2 hr on a shaker at 300 rpm. Residual proteins that were retained on the microbeads were eluted under denaturing conditions with 120 μl reducing SDS sample buffer. All protein samples were resolved on a 8% to 20% SDS-PAGE gel and analyzed by western blot. 3×HA-E6AP was probed with an anti-HA antibody at a dilution of 1:1000 overnight in a cold room, followed by incubation with IRDye^® 680RD goat anti-rabbit secondary antibody at a dilution of 1:10,000 for 30 min at room temperature. The signal of respective protein was then visualized using a LI-COR Odyssey Fc fluorescence imaging system. (B) Western blot signals were quantified densitometrically using ImageJ 1.47v, and elution efficiency was calculated as described in the protocol (see step 28). The chart was plotted with GraphPad Prism 8 version 8.4.0 (671). Native elution of 3×HA-E6AP was found to be temperature dependent. At 37°C, the efficiency decreases, indicating instability at higher temperature. To avoid instability problems, an elution temperature of 4°C was chosen for further experiments.

To identify temperature dependence on the elution efficiency, we first carried out an experiment by incubating 3×HA-E6AP bound to the anti-HA microbeads with 3×HA-peptide at 4°C, 25°C, or 37°C on a shaker at 300 rpm for 2 hr. If the elution does not depend on the elution temperature, the signal of 3×HA-E6AP should be constant at the tested temperatures, as is observed for denaturing elution. In principle, one would expect that the higher the temperature, the more the proteins are eluted because the reaction is faster at higher temperatures. We observed that the efficiency of the native elution was slightly higher when it was carried out at 25°C compared to 4°C but reduced at 37°C (Fig. 3). This indicates that the bait protein 3×HA-E6AP is not stable at 37°C under the given conditions, thus decreasing the yield of Co-IP. Furthermore, the prey protein HPV16E6 is known to be a temperature-sensitive protein. Because of these stability issues, the subsequent experiments were carried out at 4°C.

Next, the optimal concentration of peptide required for maximum elution efficiency was investigated. The higher the eluent concentration, the higher the amount of eluted protein, until a maximal efficiency is reached. This efficiency might then be limited by other factors, such as the kinetic parameter, i.e., elution time. For this test, 3×HA-E6AP bound to anti-HA microbeads was competitively displaced with increasing amounts of 3×HA-peptide (0, 50, 75, 150, 500, and 1000 μM) for 2 hr at 4°C on a shaker at 300 rpm as shown in Figure 4A. Under the given conditions, the elution efficiency was maximal at ∼50% starting at an eluent concentration of 150 μM 3×HA-peptide. This can be seen in Figure 4B, where the signals reach a plateau at ∼50%. In further experiments, we used a 250 μM eluent concentration to maintain a high yield of elution.

Figure 4

Elution efficiency with increasing concentration of 3×HA-peptide. (A) Cell lysates of HEK293T cells overexpressing 3×HA-E6AP were divided equally, corresponding to 2500 μg/ml total protein. Each was incubated with 50 μl anti-HA microbeads. The bead solution was transferred to a μColumn, a magnetic field was applied, and unbound proteins were washed away as described above (Fig. 2). 3×HA-E6AP complex was eluted with 120 μl of different concentrations of 3×HA-peptide at 4°C for 2 hr on shaker at 300 rpm. Residual proteins that were retained on the microbeads were eluted under denaturing conditions with 120 μl reducing SDS sample buffer. All protein samples were resolved on a 8% to 20% SDS-PAGE gel and analyzed by western blot. 3×HA-E6AP was probed with an anti-HA antibody at a dilution of 1:1000 overnight in a cold room, followed by incubation with IRDye^® 680RD goat anti-rabbit secondary antibody at a dilution of 1:10,000 for 30 min at room temperature. The signal of respective protein was then visualized using a LI-COR Odyssey Fc fluorescence imaging system. (B) Western blot signals were quantified densitometrically using ImageJ 1.47v, and elution efficiency was calculated as described in the protocol (see step 28). The chart was plotted with GraphPad Prism 8 version 8.4.0 (671). The amount of eluted 3×HA-E6AP increases with higher amounts (higher eluent concentrations) of 3×HA-peptide. A maximum of ∼50% was achieved starting at an eluent concentration of 150 μM 3×HA-peptide. As a consequence, a 250 μM eluent concentration was chosen for further experiments.

To investigate the time course of the performance of 3×HA-peptide elution, we incubated 3×HA-E6AP-saturated anti-HA microbeads with 250 μM 3×HA-peptide for different lengths of time, as indicated in Figure 5A, at 4°C on a shaker at 300 rpm. Over time, the amount of the eluted protein 3×HA-E6AP should increase until the reaction completes and reaches a plateau. In Figure 5B, indeed, the protein signals increased over time, indicating that the longer the incubation time, the more the 3×HA-E6AP was eluted. An overnight incubation allowed the elution of ∼60% 3×HA-E6AP. We did not want to prolong the incubation time further in order to avoid artifacts caused by protein aggregation or degradation.

Figure 5

Effect of elution time on the elution efficiency of 3×HA-E6AP. (A) Cell lysates of HEK293T cells overexpressing 3×HA-E6AP were divided equally, corresponding to 2500 μg/ml total protein. Each was incubated with 50 μl anti-HA microbeads. The bead solution was transferred to a μColumn, a magnetic field was applied, and unbound proteins were washed away as described above (Fig. 2). 3×HA-E6AP complex was eluted with 120 μl of 250 μM 3×HA-peptide at 4°C and incubated for different numbers of hours on a shaker at 300 rpm. Residual proteins that were retained on the microbeads were eluted under denaturing conditions with 120 μl reducing SDS sample buffer. All protein samples were resolved on a 8% to 20% SDS-PAGE gel and analyzed by western blot. 3×HA-E6AP was probed with an anti-HA antibody at a dilution of 1:1000 overnight in a cold room, followed by incubation with IRDye^® 680RD goat anti-rabbit secondary antibody at a dilution of 1:10,000 for 30 min at room temperature. The signal of respective protein was then visualized using a LI-COR Odyssey Fc fluorescence imaging system. (B) Western blot signals were quantified densitometrically using ImageJ 1.47v, and elution efficiency was calculated as described in the protocol (see step 28). The chart was plotted with GraphPad Prism 8 version 8.4.0 (671). The elution efficiency increases over time and is highest after overnight incubation. Consequently, an overnight incubation of the washed magnetic beads was performed in subsequent experiments.

We never reached 100% elution efficiency when applying the 3×HA-peptide. This might be for several reasons. First of all, the kinetic test did not reach a plateau and could be prolonged. However, this should be analyzed carefully because the proteins are still in a rather crude environment and prone to degradation and stability issues. Second, the bait protein might also bind nonspecifically to the bead matrix; elution with the 3×HA-peptide is simply not possible in this case.

Co-elution of prey proteins of the ternary complex

With this experiment, we wanted to address two questions: (i) Is it possible to pull down the ternary complex? (ii) Can we reduce the protein background by native elution?

• (i) It is known that HPV16E6 oncoprotein recruits E6AP to form a ternary complex with tumor suppressor p53 and leads to the ubiquitination and proteasomal degradation of p53 (Martinez-Zapien et al., 2016). This is a quite complicated complex because the binding of E6AP is required for conformational changes in 16E6 that enable binding of p53. Whether there is a direct interaction of p53 with E6AP within the complex is not known. We used this example to demonstrate that the natively eluted 3×HA-E6AP maintains its biological function during the isolation process and to show that our strategy allows us to isolate multimeric complexes of proteins even with indirect interactions, here p53 and E6AP. The bait protein 3×HA-E6AP and the viral protein 16E6 (first targeting prey protein) were heterologously expressed in the mammalian cell line HEK293T. Endogenous p53 was the second prey protein, with all three proteins together forming the ternary complex. As a control, the same experiment was conducted without co-expression of 16E6 or 3×HA-E6AP. We applied the optimized native elution strategy by competitively displacing 16E6/E6AP/p53 ternary complex with 250 μM 3×HA-peptide overnight. The input samples, which corresponded to the cell lysate, verified the expression of the proteins of interest and the equal loading of cell lysate on the magnetic beads, as shown by the similar signal of endogenous GAPDH as a loading control (input, Fig. 6A). Regarding the controls, 3×HA-E6AP in the absence of 16E6 did not pull down p53. The second control, the heterologous expression of 16E6 only, showed a signal neither for p53 nor for 16E6 in the natively eluted samples. This meets expectations because there was no HA-tagged protein to be pulled down with the anti-HA magnetic beads, and this verifies that there was no unspecific signal caused by unspecific binding of p53 or 16E6. Expressing 3×HA-E6AP together with 16E6, we could pull down the ternary complex from crude cell lysate, demonstrated by the clear signals for p53 and 16E6 in the western blots both in the native elution and under denaturing elution (Fig. 6A). This means that we could pull down p53 as an indirect binder of E6AP.
• (ii) Western blotting only allows the visualization of proteins specifically recognized by the applied antibodies. In order to analyze the entire protein content of the sample, meaning the purity of the sample and detection of nonspecific binders, we additionally performed Coomassie staining of a reducing SDS-PAGE gel (Fig. 6B) in analogy to the western blot. Here, we could detect a protein band corresponding to 3×HA-E6AP. The sensitivity of the Coomassie staining was too low to detect 16E6 or p53. However, higher purity and less background were clearly visible in protein samples natively eluted with 3×HA-peptide compared with the residual proteins that were eluted under denaturing conditions. All residual proteins that are commonly co-eluted under denaturing conditions gave a high background in further analysis, e.g., mass spectrometry, which is often applied after Co-IP. One of the common problems in pull-down assays is the co-elution of IgG antibodies or the heavy and light chains of antibodies, which leads to high background or smearing for the pulled-down proteins. Protein bands of heavy and light chains from the antibody could hamper detection of the signal of bait or prey of interest if they have similar molecular weights (∼50 kDa for heavy chain and ∼25 kDa for light chain, indicated with * in Fig. 6A). Given that the anti-HA antibody immobilized on the microbeads is from mouse, incubation of the blot membrane with goat anti-mouse secondary antibody will detect these antibodies. In native elution, no signals for co-eluted antibody light chains were detected. This clearly shows that this common issue could be solved easily by applying native elution. Furthermore, in the denaturing elution of residual proteins, a lot of E6 remained bound to the column, which can also be seen for the control sample E6 only (Fig. 6B). This indicates that E6 binds nonspecifically to the anti-HA microbeads. We did not observe a signal for E6 in the native elution of controls, indicating that this elution strategy could overcome the unspecific binding of the prey.

Figure 6

Co-immunoprecipitation of 3×HA-E6AP/16E6/p53 ternary complex under native conditions. (A) Cells overexpressing 3×HA-E6AP or 16E6 or both, with a total protein concentration of 2500 μg/ml, were prepared. Each was incubated with 50 μl anti-HA microbeads. The bead solution was transferred to a μColumn, a magnetic field was applied, and unbound proteins were washed away as described above (Fig. 2). 3×HA-E6AP that was bound on anti-HA microbeads was incubated with 120 μl of 250 μM 3×HA-peptide overnight at 4°C with gentle shaking at 300 rpm. Residual proteins that were retained on the microbeads were eluted under denaturing conditions with 120 μl reducing SDS sample buffer. All protein samples were resolved on a 8% to 20% SDS-PAGE gel and analyzed by western blot. Proteins of interest were bound to respective antibodies overnight in a cold room: anti-HA antibody at a dilution of 1:1000, anti-p53 (DO-1) antibody at a dilution of 1:1000, anti-GAPDH antibody at a dilution of 1:500 as a loading control, or anti-16E6 antibody at a dilution of 1:10,000. This was followed by incubation with IRDye^® 680RD goat anti-rabbit secondary antibody or IRDye^® 680RD goat anti-mouse secondary antibody at a dilution of 1:10,000 for 30 min at room temperature. The signal of respective protein was then visualized using a LI-COR Odyssey Fc fluorescence imaging system. The input samples resembled the cell lysate loaded on the magnetic beads, and signals here verified the expression of the proteins of interest 3×HA-E6AP, p53, and 16E6. The native and denaturing elutions showed that 3×HA-E6AP alone cannot pull down p53 and that 16E6 and p53 in the absence of 3×HA-E6AP do not bind and elute nonspecifically. In the presence of 3×HA-E6AP and 16E6, a co-elution of p53 and 16E6 occurs, as demonstrated by the signals for all three proteins in the native and denaturing elutions. (B) The samples from (A) were resolved on another 8% to 20% SDS-PAGE gel, followed by Coomassie R-250 staining in order to analyze the total protein content. Lysates of non-transfected cells were additionally applied to the SDS-PAGE gel and resembled nonspecific binders only (without heterologous 3×HA-E6AP or 16E6). The input samples resembled the overall protein content loaded on the magnetic beads. The samples of the native elution contained significantly less overall protein content compared to the denaturing elution, i.e., the native elution with 3×HA-peptide was able to avoid elution of almost all nonspecifically bound proteins as well as the light chain of anti-HA antibody [indicated with * in (A)].

These experiments showed that this method is useful in preparing functional protein samples for protein-protein interaction studies.

Variations

The isolation procedure for the antibody-bait/prey complex can alternatively be performed by using agarose beads or Protein A/G beads.

The same strategy could also be applied when other short tags, such as V5 or FLAG, are fused to the bait protein. Native elution with commercially available single or triple FLAG peptide was reported before (see Current Protocols article; Brizzard & Chubet, 1997; Hernan, Heuermann, & Brizzard, 2000).

Time Considerations

Cell culturing, Co-IP, and overnight incubation with the peptide for native elution take 2 days. Western blot analysis takes another 1 day. In sum, the entire protocol takes 3 days. Of course, individual applications, e.g., with other cell lines or antibodies with different optimal requirements, can shorten or prolong this experiment. Notably, preparing the 3×HA-peptide stock solution can take 1 to 2 hr because the pH needs to be carefully titrated to neutral pH.

Acknowledgments

We would like to acknowledge the Wilhelm Sander-Stiftung for funding. We are grateful to Dr. Murielle Masson for providing us with the HEK293T cells. Anti-HPV16E6 antibody was generously provided by Arbor Vita Corporation.

Open access funding enabled and organized by Projekt DEAL.

Author Contributions

JiaWen Lim: Data curation; Methodology; writing-original draft. Thomas Iftner: Funding acquisition; Supervision. Claudia Simon: Conceptualization; methodology; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Software; Supervision; Validation; Visualization; writing-original draft; writing-review & editing.

Conflict of Interest

The authors declare no conflict of interest.

Open Research

Data Availability Statement

Data sharing not applicable–no new data generated.

Corrections

In this publication, author's conflict of interest and data availability statement have been added.

The current version online now includes this information and may be considered the authoritative version of record.