Chapter 20
Protein–Protein Interactions: Pull-Down Assays
Arthur Louche, Suzana P. Salcedo, and Sarah Bigot
Abstract
Determining protein partners is an essential step toward understanding protein function and identifying
relevant biological pathways. Many methods exist for investigating protein–protein interactions. The pull-
down assay is an in vitro technique used to detect physical interactions between two or more proteins and
an invaluable tool for confirming a predicted protein–protein interaction or identifying novel interacting
partners. This method typically involves the use of affinity purification with various wash and elution steps.
In this chapter, we describe how an interaction between two purified bacterial proteins or between bacte-
rial and eukaryotic proteins can be detected by pull-down experiments.
Key words Pull-down, Protein–protein interactions, Tagged protein, Affinity purification
1 Introduction
Pathogenic bacteria produce virulence factors that usually help the
pathogen to survive in an environmental niche, to promote coloni-
zation and invasion of host tissues, or to modulate the immune
system. Virulence factors are toxins or effector proteins than can be
transported by diverse secretion machineries in bacteria [1, 2].
Once secreted, these proteins can be assembled on the bacterial
cell surface, released in the extracellular space, or secreted directly
into a host cell or a neighboring bacterium. Once in host cells,
effectors often target key proteins to hijack the host cellular
machinery to remodel signaling cascades. The yeast two-hybrid
system is often used to screen a large number of host proteins that
potentially interact with bacterial effectors [3]. Regarding the
mechanism of the secretion systems, a bacterial two-hybrid system
is frequently employed to identify interaction networks between
components of the secretory apparatus, as well as interaction
between effectors and proteins of the machinery [4]. However,
protein–protein interactions that have been determined by two-
hybrid assay must be confirmed by other methods [5].
Laure Journet and Eric Cascales (eds.), Bacterial Protein Secretion Systems: Methods and Protocols, Methods in Molecular
Biology, vol. 1615, DOI 10.1007/978-1-4939-7033-9_20, © Springer Science+Business Media LLC 2017
247
248 Arthur Louche et al.
Pull-down is an in vitro method widely used to detect or con-
firm interactions among multiple proteins. This assay is similar in
methodology to co-immunoprecipitation experiments in its use of
an affinity ligand to capture interacting proteins. The difference
between these two methods is that while co-immunoprecipitation
uses immobilized antibodies to capture protein complexes, the
pull-down approach uses a purified and tagged protein as a “bait”
to bind any interacting proteins. The method consists of first
immobilizing the tagged protein (bait) on an affinity ligand specific
to the tag, creating an affinity support to capture and purify other
proteins (prey) that interact with the bait. The bait and prey pro-
teins can be obtained from multiple sources, such as cell lysates,
purified proteins, expression systems, and in vitro transcription/
translation systems. Once the prey proteins have been incubated
with an immobilized bait protein, interacting complexes are eluted
using an eluting buffer depending on the affinity ligand. Each
experiment needs proper controls to demonstrate that character-
ized interactions are not an artifact. For example, a positive control
consisting of an immobilized bait protein alone is necessary to
verify proper attachment of the tagged bait protein to the affinity
support. To identify and eliminate false positives caused by nonspe-
cific binding of prey proteins to the affinity support, cell lysates or
purified proteins can be analyzed after being passed through a
minus bait support. Following a pull-down experiment, protein
fractions are resolved by sodium dodecyl sulfate-polyacrylamide
gel electrophoresis (SDS-PAGE) and then visualized by gel stain-
ing or western-blotting detection.
In this chapter, we describe detailed pull-down assay proce-
dures that allow the identification of interacting proteins. First, we
focus on how to perform a pull-down experiment to identify an
interaction between a bacterial bait protein and eukaryotic prey
proteins expressed in host cells (Subheadings 3.1 and 3.2). Next,
we present how the interaction between two purified proteins can
be visualized by a pull-down assay (Subheading 3.3). In these pro-
cedures, pull-down experiments have been performed using spe-
cific bait proteins fused to a 6× histidine tag. As a consequence, we
selected Ni-NTA agarose beads as the affinity support used to
immobilize these recombinant proteins.
2 Materials
Prepare all solutions with distilled water at room temperature and
keep them at the indicated temperatures.
2.1 Preparation 1. Eukaryotic cells.
of Cell Lysate 2. Cell culture dish, treated for optimal cell attachment, with
growth surface area around 55 cm2, sterile.
Pull-Down Assays 249
3. Plasmid containing the gene of interest fused to a specific tag
(obtained from a EndoFree maxipreparation).
4. Transfection reagent.
5. Phosphate buffered saline (PBS): Prepare a 10× solution with
bidistilled water (18.2 MΩ cm) containing 10.6 mM KH2PO4,
30 mM Na2HPO4, 2H2O, and 1.54 M NaCl, and sterilize with
a 0.2 μm filter. The 1× solution obtained following dilution
with bidistilled water will have a pH of around 7.4.
6. Radioimmunoprecipitation assay (RIPA) buffer: Ready-to-use
solution containing 150 mM NaCl, 1.0% IGEPAL® CA-630,
0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris, pH 8.0.
7. Antiprotease cocktail: Mix 1% (v/v) of protease inhibitor cock-
tail (Sigma-Aldrich), phosphatase inhibitor cocktail 2 (Sigma-
Aldrich), phosphatase inhibitor cocktail 3 (Sigma-Aldrich),
and phenylmethylsulfonyl fluoride (PMSF).
2.2 Pull- 1. 1 M Tris–HCl, pH 7.5 stock solution. Weigh 121.1 g Tris base
Down Assays and transfer to a 1 L graduated cylinder. Add water to 800 mL,
mix, adjust pH with HCl, and make up to 1 L with water.
Store at room temperature (see Note 1).
2. 5 M NaCl stock solution. Weigh 292.2 g NaCl and transfer to
a 1 L graduated cylinder. Add water to 800 mL, stir, and adjust
volume to 1 L with water (see Note 1).
3. Equilibrium buffer (see Note 2): 20 mM Tris–HCl, pH 7.5,
250 mM NaCl. Mix 1 mL 1 M Tris–HCl, pH 7.5 stock solu-
tion with 2.5 mL 5 M NaCl stock solution in a 50 mL centri-
fuge tube, and add water to a volume of 50 mL. Keep at 4 °C
(see Note 3).
4. Elution buffer (see Note 2): 20 mM Tris–HCl, pH 7.5, 250
mM NaCl, 500 mM imidazole. Weigh 1.7 g imidazole in 50
mL solution of equilibrium buffer. Keep at 4 °C (see Note 3).
5. Purified His-tagged protein (bait).
6. Ni-NTA agarose beads: 6% beaded agarose (cross-linked), pre-
charged with Ni2+ (Protino® Ni-NTA Agarose, Macherey
Nagel, or equivalent). Store at 4 °C (see Note 4).
7. 0.8 mL empty columns for gravity flow (Pierce™ Centrifuge
Columns, Thermo Fisher Scientific, or equivalent).
8. Refrigerated microcentrifuge.
2.3 Sodium Dodecyl 1. Resolving gel: 1.5 M Tris–HCl, pH 8.8. Weigh 90.8 g, transfer
Sulfate (SDS) to 500 mL graduated cylinder, and add 300 mL water. Adjust
Polyacrylamide Gel pH with HCl and fill with water to 500 mL. Store at room
Components temperature.
2. Stacking gel buffer: 0.5 M Tris–HCl, pH 6.8. Weigh 30.275 g,
transfer to 500 mL graduated cylinder, and add 300 mL water.
250 Arthur Louche et al.
Adjust pH with HCl and fill with water to 500 mL. Store at
room temperature.
3. 30% acrylamide/Bis solution (37.5:1 acrylamide:Bis). Store at
4 °C.
4. Ammonium persulfate (APS): 20% solution in water. Store at
−20 °C (see Note 5).
5. N,N,N′,N′-tétraméthyléthylènediamine (TEMED). Store at
room temperature.
6. SDS-PAGE running buffer: 25 mM Tris–HCl, 192 mM gly-
cine, 0.1% SDS. Prepare 10× running buffer solution: Weigh
30 g Tris base, 144 g glycine, and 10 g SDS and add distilled
water to 1 L. Store at room temperature. Prepare fresh 1×
solution before gel electrophoresis.
7. Laemmli lysis buffer [6], 4× concentrate: 62.5 mM Tris–HCl
pH 6.8, 2% SDS, 10% glycerol, 0.01% bromophenol blue, 5%
β-mercaptoethanol. Store at −20 °C (see Note 6).
8. Protein ladder.
3 Methods
3.1 Preparation 1. Seed eukaryotic cells at 5.105 in 10 cm cell culture dish
of Cell Lysate (see Note 7) and incubate overnight at 37 °C in CO2.
2. Transfect cells with plasmid containing gene of interest fused
to a specific tag with appropriate transfection reagent for time
necessary for optimal expression of protein (16–24 h is usually
a good range).
3. Cool cells by placing plates on ice, wash cells with 1× PBS. Add
2 mL cold PBS and harvest cells using cell scraper.
4. Centrifuge 5 min at 80 × g at 4 °C.
5. Resuspend cells with 200 μL RIPA buffer supplemented with
antiprotease cocktail.
6. Incubate on ice 20 min and mix gently every 5 min with a
P200 micropipette.
7. Stock prepared cells at −80 °C (see Note 8).
8. Right before pull-down experiment, thaw prepared cell extract.
Centrifuge at 17,000 × g at 4 °C for 20 min. Use the superna-
tant as prey by following step 9 in Subheading 3.2 (see
Note 9).
3.2 Pull-Down Assay 1. Transfer 120 μL Ni-NTA agarose beads to gravity flow column
Using Cell Lysate (see Note 12).
as Prey (See Notes 10 2. Centrifuge column for 1 min at 1000 × g at 4 °C. Discard
and 11) flow-through.
Pull-Down Assays 251
3. Add 400 μL distilled water to column (see Note 13).
4. Centrifuge column for 1 min at 1000 × g at 4 °C. Discard
flow-through.
5. Mix carefully 50 μg His-tagged protein (bait) with 400 μL equi-
librium buffer and load onto column (see Notes 14 and 15).
6. Incubate 1 h (see Note 16) with agitation at 4 °C (see Note
17) and 10 min on ice without agitation (see Note 18).
7. Centrifuge column for 1 min at 1000 × g at 4 °C and keep
flow-through.
8. Load flow-through to column, and centrifuge column for
1 min at 1000 × g at 4 °C (see Note 19). Keep flow-through at
4 °C for analysis.
9. Mix 200 μL cell extract (see Note 20) with 200 μL equilibrium
buffer and load onto column (see Note 21).
10. Incubate 1 h at 4 °C under agitation (see Note 22) then 10 min
on ice without agitation (see Note 18).
11. Centrifuge column for 1 min at 1000 × g at 4 °C. Keep flow-
through for analysis.
12. Wash column by adding to column 400 μL equilibrium
buffer.
13. Centrifuge column for 1 min at 1000 × g at 4 °C. Discard
flow-through.
14. Wash column by adding to column 400 μL equilibrium buffer
containing 50 mM imidazole. Keep the first washing for
analysis.
15. Centrifuge column for 1 min at 1000 × g at 4 °C. Discard
flow-through.
16. Repeat steps 14 and 15 three times and go to step 17. Keep
last washing fraction at 4 °C for analysis.
17. Elute by loading 80 μL elution buffer to column and incubate
10 min at 4 °C (see Note 18).
18. Centrifuge column for 1 min at 1000 × g at 4 °C and keep
eluted fraction.
19. Repeat steps 17 and 18 with eluted fraction (see Note 22).
Keep eluted fraction at 4 °C for analysis.
3.3 Pull-Down Assay 1. Incubate 50 μg His-tagged bait protein with 50 μg purified prey
Using Purified Protein protein in total volume of 400 μL equilibrium buffer (see Note
as Prey (See Note 11) 23) 2 h 30 min at 4 °C under agitation (see Notes 17 and 24).
2. Add 80 μL Ni-NTA agarose beads to gravity flow column and
follow steps 1–4 of Subheading 3.2.
252 Arthur Louche et al.
3. Equilibrate column by adding 400 μL equilibrium buffer sup-
plemented with 20 mM imidazole.
4. Centrifuge column for 1 min at 1000 × g at 4 °C. Discard
flow-through.
5. Load 400 μL incubated bait and prey proteins onto column.
Incubate 10 min on ice without agitation (see Note 18).
6. Centrifuge column for 1 min at 1000 × g at 4 °C. Keep flow-
through at 4 °C for analysis.
7. Wash by adding to column 400 μL equilibrium buffer supple-
mented with 20 mM imidazole.
8. Centrifuge column for 1 min at 1000 × g at 4 °C. Save first
washing at 4 °C for analysis.
9. Repeat washing steps 7 and 8 four times and keep last washing
fraction at 4 °C for analysis.
10. Add 200 μL elution buffer to column and incubate on ice
10 min.
11. Centrifuge column for 1 min at 1000 × g at 4 °C. Keep eluted
fraction at 4 °C for analysis.
3.4 SDS-PAGE 1. To 15 μL protein fraction add 5 μL Laemmli lysis buffer, 4×
and Analysis concentrate. Heat for 3 min at 100 °C and centrifuge 30 s
of Protein Fractions using a microcentrifuge to bring down condensate.
2. Load 10 μL protein fraction and 5 μL protein ladder on SDS-
polyacrylamide gel.
3. Electrophorese proteins in running buffer at 100 V for 15 min
then 180 V until dye front has reached bottom of gel.
4. Identify interacting proteins by immunodetection or blue coo-
massie coloration (see Note 25).
4 Notes
1. We prefer not to use the solutions after 6 months of storage.
2. A different buffer, such as HEPES (4-(2-hydroxyethyl)-1-
piperazineethanesulfonic acid), MES (2-(N-morpholino) eth-
anesulfonic acid), or phosphate buffers, may be required for
your specific protein–protein interaction. Additionally,
different pH values may be tested as these are specific and
dependent on the interaction between proteins.
3. We found that pull-down experiments work better with fresh
equilibrium and elution buffers.
4. The bait proteins used in this protocol are tagged with 6× His
that bind the nickel agarose affinity support. The choice of the
matrix-associated antibody depends on the fusion tag. The His
Pull-Down Assays 253
tag is composed of a peptide motif that consists of six histidine
residues with a high affinity towards metals like nickel that
composes the used Ni-NTA agarose but also the Ni-NDA,
Ni-TED, or Ni-TALON resins. The 6× His tag is very small
(~1 kDa), which renders it less immunogenic than other larger
tags, is shown not to affect the native conformation of bait
proteins, and maintains its partner binding activity. Few natu-
rally occurring proteins also bind to Ni-NTA matrices, making
this tag the most commonly used affinity tag. In pull-down
assays, the choice of the matrix-associated antibody depends
on the fusion tag. What follow are some examples of tags with
their advantages and disadvantages. The FLAG tag is an octa-
peptide that is likely located on the surface of the fusion pro-
tein due to the hydrophilic nature of amino acid residues and
has affinity to anti-FLAG resin. Like the His tag, the FLAG tag
is small, but a disadvantage is that the monoclonal antibody
matrix is not as stable as Ni-NTA. Glutathione S-transferase
(GST) tag binds to glutathione-associated support with high
affinity and specificity. This tag has the advantage that GST
isoforms are not normally found in bacteria, so purified bacte-
rial prey proteins normally do not have affinity with glutathi-
one resin. However, GST tag is large (26 kDa), exists as a
dimer, is prone to nonspecific interaction, is expensive, and
affinity to its support depends on certain reagents. The malt-
ose-binding protein (MBP) tag from an Escherichia coli peri-
plasmic protein has affinity for matrix consisting of sugars or
anti-MBP. This tag is used for the purposes of overcoming
problems associated with the expression and purification of
recombinant proteins [7]. However, the disadvantage of the
MBP tag is its large size, its immunogenicity, and the mild elu-
tion of MBP-tagged proteins, which complicate pull- down
experiments.
5. Make an aliquot of 1 mL before −20 °C storage. This will pre-
vent the degradation caused by repeated thawing.
6. Make an aliquot of 500 μL before −20 °C storage. The used
Laemmli lysis buffer can be kept at 4 °C for 1 month.
7. As negative control, prepare a cell lysate without expressing
bait protein (negative cell lysate). This will eliminate false posi-
tives resulting from nonspecific interactions of cell lysate
proteins with the Ni-NTA agarose beads. Additional negative
controls can include an irrelevant protein with the same tag or
expression of the tag alone, as in the case of the GFP.
8. Before stocking the cells, remove an aliquot and control by
western blot the production of the prey protein.
9. Whole-cell lysate instead of the supernatant fraction can also
be used to test whether the prey protein of interest localizes in
the pellet fraction.
254 Arthur Louche et al.
10. Pull-down experiments using cell lysates will not demonstrate
that interaction between the bait and prey proteins is direct but
only determine that they are part of the same complex. To
prove a direct interaction, the prey protein must be purified
and used in pull-down experiments as described in Subheading
3.3.
11. Try to work mostly on ice or at 4 °C to prevent the degrada-
tion or the denaturation of the proteins.
12. Break the end cap of the gravity flow column and place it on a
1.5 mL Eppendorf tube. Thoroughly resuspend the Ni-NTA
resin by inverting the bottle several times to obtain a uniform
suspension. Pipette tips must be cut to allow the Ni-NTA aga-
rose beads to get into.
13. This step eliminates the left 30% ethanol present in the Ni-
NTA resin.
14. Before loading the bait protein, plug the gravity flow column
using a piece of parafilm before replacing it on a 2 mL
Eppendorf tube.
15. Prepare a supplementary column by mixing 50 μg of a known
noninteracting bait fused to 6× His tag with 400 μL equilib-
rium buffer to an empty column. Additionally, prepare a col-
umn by adding 400 μL equilibrium buffer to an empty column.
These negative bait columns will be used in combination with
cell lysates to eliminate false positives resulting from nonspe-
cific interactions.
16. The incubation time can be increased from several hours to
overnight at 4 °C under agitation depending on the strength
of the interaction between bait and prey proteins.
17. Rotate on roller or rotating platform.
18. The column should stand straight on the ice. This step allows
the resin to flow by gravity before centrifugation.
19. We found that loading two times the flow-through increases
the capacity of the binding.
20. The volume is dependent on the protein concentration of the
cell extract. As a guide, 125–150 μg of protein of a cell extract
is usually incubated per microgram of bait protein. A
lternatively,
cell extract samples can be normalized by visualization of trans-
fected proteins to ensure equivalent expression of the prey and
the relevant controls (see Note 7).
21. Several controls should be added at this step. Load 400 μL
equilibrium buffer without prey protein to analyze the effi-
ciency of the immobilization of the bait protein. As negative
controls, load onto the negative column (see Note 12) 200 μL
cell lysate containing the prey protein or the negative cell lysate
(see Note 7) mixed with 200 μL equilibrium buffer. Additionally,
Pull-Down Assays 255
load 200 μL negative cell lysate mixed with 200 μL equilib-
rium buffer onto the column associated with the bait protein.
22. We found that loading two times the eluted fraction increased
its quantity.
23. As negative control, incubate 50 μg bait protein (minus prey)
or prey protein alone (minus bait) in 400 μL equilibrium buf-
fer. The minus prey control will ensure that the Ni-NTA aga-
rose resin will correctly capture the His-tagged bait protein
alone. The minus bait control will eliminate false positives
resulting from an interaction between affinity support and prey
protein.
24. A different incubation temperature and time may be required
for your specific protein–protein interaction.
25. A prey protein that interacts with the bait protein will be found
in the eluted fraction. In contrast, a noninteracting protein will
not be retained by the bait protein, will pass through the col-
umn, and will be found in the flow-through protein fraction.
References
1. Costa TRD, Felisberto-Rodrigues C, Meir A, Cascales E (2014) Architecture and assembly of
Prevost MS, Redzej A, Trokter M, Waksman the Type VI secretion system. Biochim Biophys
G (2015) Secretion systems in Gram-negative Acta 1843:1664–1673
bacteria: structural and mechanistic insights. 5. Boucrot E, Henry T, Borg J-P, Gorvel J-P,
Nat Rev Microbiol 13:343–359 Méresse S (2005) The intracellular fate of
2. McBride MJ, Nakane D (2015) Flavobacterium Salmonella depends on the recruitment of
gliding motility and the type IX secretion sys- kinesin. Science 308:1174–1178
tem. Curr Opin Microbiol 28:72–77 6. Laemmli UK (1970) Cleavage of structural
3. Rodríguez-Negrete E, Bejarano ER, Castillo proteins during the assembly of the head of
AG (2014) Using the yeast two-hybrid sys- bacteriophage T4. Nature 227:680–685
tem to identify protein-protein interactions. 7. Di Guan C, Li P, Riggs PD, Inouye H (1988) Vectors
Methods Mol Biol 1072:241–258 that facilitate the expression and purification
4. Zoued A, Brunet YR, Durand E, Aschtgen M-S, of foreign peptides in Escherichia coli by fusion
Logger L, Douzi B, Journet L, Cambillau C, to maltose-binding protein. Gene 67:21–30
