Introduction
GTPases cycle from inactive (GDP-bound)
to active (GTP-bound) forms that interact with and regulate components
of intracellular signaling pathways. This cycle is under the control of
several classes of regulatory protein. Under basal conditions, the Rho
GTPases are complexed with their cytosolic GDP-Dissociation Inhibitor
(GDI), which stabilizes the inactive GDP-bound form. Upon cell
stimulation, the GTPase dissociates from the GDI and translocates to
the plasma membrane. Subsequently the GTPase releases GDP and binds
GTP, a reaction promoted by Guanine nucleotide Exchange Factors (GEFs),
which leads to the interaction with specific molecular targets.
Inactivation of the GTPase results from hydrolysis of GTP into GDP, and
the intrinsic rate of hydrolysis by the GTPase is enhanced by the
action of GTPase-Activating Proteins (GAPs). The active form of the
GTPase appears to be transient and labile, which makes its measurement
difficult. Indirect methods based upon monitoring association with or
activation of regulatory proteins or upon relocalization within the
cell have been used to assess Rho GTPase activation during the dynamic
cell response. These methods have their problems and limitations, and
to really characterize Rho GTPase activation a direct measurement of
the formation of the active form within the cell is necessary. A
classic method to analyze direct GTPase activation consists in
measuring the GTP:GDP ratio after 32P labeling of cells and
immunoprecipitation of the GTPase. In addition to the need for high
levels of radioactivity, this approach is limited for Rho GTPases by
the lack of efficient immunoprecipitating antibodies and the rapid
turnover of GTP-bound state. More recently a new assay measuring Rho
GTPase activation has been developed based on the fact that only the
active form of the GTPase interacts with downstream effectors.
Identification of the GTPase-binding region on target proteins, and the
expression of these binding domains in active soluble forms allows them
to be used as specific probes to detect the active form of the GTPase.
Principle of the assay
The Rac and Cdc42 (p21)-activated kinase
PAK1 contains in the N-terminal regulatory region a specific site for
the interaction with the active form of these two GTPases. This region,
referred to as the CRIB domain (Cdc42/Rac Interactive Binding domain)
or p21-binding domain (PBD), was first described by Burbelo et al.
1 and the minimal consensus sequence identified for specific
GTPase binding corresponds to amino acids 74 to 89 in PAK1. A
homologous, but Cdc42-selective, CRIB domain is also found in the
Wiskott Aldrich Syndrome Protein (WASP), amino acids 235 to 268, which
shares 50% homology with the PAK CRIB domain. Binding affinities of the
PAK PBD range from ~20 nM to ~1 M, depending on the length of the
peptide encompassing the CRIB domain used.2 The affinity
precipitation assay detecting active Rac and Cdc42 uses the p21-Binding
Domain (PBD) of the target PAK1 fused to glutathione S-transferase
(GST) to isolate (precipitate) a complex containing the GST-PBD bound
to the active GTPase (Rac-GTP or Cdc42-GTP). The precipitated GTPase is
then quantified by Western blotting using specific antibodies (Fig.1).
Method for GST-PBD preparation
Construction of a GST-PAK1 PBD fusion protein
The cDNA of the coding sequence for PAK1 PBD corresponding to amino acids 67 to 150 was amplified by PCR, cloned into a pGEX-2T vector at the BamH1 / EcoR1 sites, and used to transform DH-10B strain of E.coli. The bacteria then produces a fusion protein with glutathione S-transferase (GST) upstream of the PBD. Transformed bacteria are stored at -80°C in 20% glycerol L-broth medium.
Expression and purification of the fusion protein
30 ml of L-broth medium containing 100 g/ml ampicilin is inoculated with the glycerol stock and grown overnight at 37°C. The overnight culture is used to inoculate 1 L broth medium/100 g/ml ampicilin and is grown for 3 hours at 37°C. Transcription is then induced by adding 0.8 mM isopropyl--D-thiogalactopyranoside (IPTG) for 3 hours at 30°C. The bacterial suspension is centrifuged for 10 minutes at 5000 rpm and resuspended in 10 ml of lysis buffer (50 mM Tris-HCl pH7.5, 150 mM NaCl, 5 mM MgCl2, 1 mM dithiothreitol, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mg/ml lysozyme, 20 g/ml DNAse I and 1 g/ml aprotinin). The E.coli lysate is incubated 30 minutes on ice, then sonicated and centrifuged for 10 minutes at 10,000 rpm at 4°C.
The fusion protein is purified from the supernatant with glutathione-sepharose 4B beads (Pharmacia) previously washed 5 times in lysis buffer. Glutathione-sepharose beads are incubated with the supernatant for 2 hours at 4°C, then the beads are centrifuged 5 minutes at 2000 rpm at 4°C and washed 5 times with washing buffer (50 mM Tris-HCl pH8, 150 mM NaCl, 5 mM MgC12, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin). Beads can be aliquoted at this point in washing buffer with 10% glycerol and stored at -80°C. The integrity of the GST-PBD (MW = ~36 kDa) is checked by SDS-PAGE (Fig.2) and the final protein concentration determined. The presence of a minor fraction of PBD breakdown products is typical in these preparations, and these do not affect PBD activity.
The GST-PAK1 PBD specifically interacts with the active conformation of Rac or Cdc42, but not Rho. This should be verified after each GST-PBD purification by using recombinant GTPases loaded with GDP as a negative control or with GTPS as a positive control (Fig.3A). Accordingly, recombinant GTPases (50-100 ng) are diluted in TEDM buffer (25 mM Tris-HCl pH7.5, 1 mM EDTA, 1 mM DTT and 5mM MgCl2) and 26.4 l of the GTPase is mixed with 15 l of 0.1 M EDTA, 15 l of 1 mM GTPS or 15 l of 10 mM GDP and incubated 10 minutes at 30°C. The exchange reaction is stopped by the addition of 4 l of 1M MgCl2. Another control is to use expressed mutants of Rac or Cdc42: dominant active mutants like G12V or Q61L maintain bound GTP and should bind to the GST-PBD, while the predominantely GDP-associated dominant negative T17N mutant does not (Fig.3B).
Simple control experiments can be
performed by generating GDP-bound GTPase and GTP-bound GTPase in
vitro from the cell lysates to be used in your studies. Indeed,
this is an effective means to assess the total amount of activatable
GTPase present in your sample. Either expressed recombinant GTPases or
endogenous GTPases can be loaded with nucleotides in this manner.
Briefly, the lysate should be incubated with nucleotides in excess
(100-200 M GTPS, a non-hydrolyzable GTP analog, or 1 mM GDP) in a high
concentration of EDTA (10-20 mM) and a low concentration of Mg++
(5 mM maximum) at 30°C to induce the binding of the nucleotide.
Subsequently increasing the Mg++ concentration (to 50-70 mM)
will stop the nucleotide exchange. Specific conditions are as follows:
100 l of cell lysate containing the GTPase is mixed with 12 l of 0.2 M
EDTA and 12 l of 1 mM GTPS or 12 l of 10 mM GDP and incubated 10
minutes at 30°C. The exchange reaction is stopped by the addition
of 8 l of 1M MgCl2. Loaded GTPases should be used
immediately in the affinity precipitation assay.
Preparation of cell extract to measure GTP-bound GTPase
This step represents a critical point of the assay and determination of the most appropriate lysis buffer has to be established prior any affinity precipitation. Cells are pretreated or stimulated to generate the active form of the GTPase. For cells in suspension the stimulation is stopped by addition of an equal volume of cold 2X lysis buffer. For adherent cells, the stimulation medium is removed, cold 1X lysis buffer is added, and the cells are scraped from the plates while on ice. The resulting cell lysates are kept on ice for several minutes, then clarified by centrifugation at 4°C for 10 minutes at 10,000 rpm in a table-top microcentrifuge. Cell lysate can be used directly for the affinity precipitation assay, or can also be frozen in liquid nitrogen and stored at -80°C until use. For each affinity precipitation sample, an equal number of cells or the same amount of total cell protein should be mixed with the GST-PBD.
Detergents are necessarily used in the lysis buffer for cell extracts in order to disrupt cells and solubilize membrane proteins. Each cell type has a different detergent requirement to obtain a good protein solubilization without disruption of the nucleus. In neutrophils and breast cancer cell lines, we have used the following lysis buffer: 50 mM Tris-HCl pH7.5, 10 mM MgC12, 200 mM NaCl, 1% Nonidet P-40, 5% glycerol, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml leupeptin, and 1 g/ml aprotinin.5,6
Detergents are also present in the
binding buffer for the affinity precipitation. In general, detergents
help to prevent nonspecific binding to the GST-PBD; however, some
detergents can also disrupt specific binding. We have observed that the
specific binding of GST-PAK1 PBD to active Rac and Cdc42 is decreased
in the presence of 1% CHAPS, 1% sodium deoxycholate, or 0.1% SDS, but
is not affected by 1% Triton X-100 or 1% Nonidet P-40. Analysis of the
effect of detergents and detergent concentration on specific vs
nonspecific binding is recommended when setting up the PBD assay.
The affinity precipitation assay for Rac and Cdc42
Cell extract or recombinant GTPases are incubated with GST-PBD in a maximal final volume of 500 l for 1 hour at 4°C. If necessary the solution containing the GTPase is diluted in binding buffer (25 mM Tris-HCl pH7.5, 1 mM DTT, 30 mM MgCl2, 40 mM NaCl, 0.5% Nonidet P-40, 1 g/ml leupeptin, 1 g/ml aprotinin and 1 mM PMSF) to have an equal volume in each sample. The bead pellet is then centrifuged for 2 minutes at 2,000 rpm, 4°C and washed 3 times with washing buffer (25 mM Tris-HCl pH7.6, 1 mM DTT, 30 mM MgCl2, 40 mM NaCl, 1% Nonidet P-40, plus antiproteases) and 2 times with the same buffer without Nonidet P-40. The bead pellet is finally suspended in 20 l of Laemmli sample buffer. Proteins bound to the beads are separated on a 12% SDS-PAGE, transferred onto nitrocellulose membrane and blotted for the appropriate GTPase using specific antibodies. The exact blotting conditions should be optimized depending upon the antibody used for detection; blots can be quantified by phosphoimager or densitometry. Blots can be stripped and re-probed for Rac or Cdc42. Typical results obtained with the PAK1 PBD assay can be found in 5,6,8,9.
Ratio of GST-PBD and GTPase
In order to recover 80% of recombinant Rac1 loaded with GTPS, we determined that the amount of GST-PBD added needs to be 20 times the amount of the GTPase (Fig.4A). When starting with cell extract expressing endogenous levels of GTP-bound GTPase, 106 to 107 cells or 300 to 800 g of total protein was necessary to observe a signal in an affinity precipitation assay performed with 10 to 15 g of GST-PBD. Variations are probable depending on the cell type or the agonist used to stimulate the GTPase pathway. Since the amount of activated (GTP-bound) GTPase usually only represents a small fraction of the total available GTPase, we recommend that Western blots initially be performed on the total cell lysates and PBD-precipitated GTPS-loaded lysates in order to assess the level of signal detectable with the antibodies to be used for detection of the bound (active) GTPase.
Time-course of the interaction
The binding of GTPase to GST-PBD already coupled to the sepharose beads is rapid and reaches about 75% of the maximum after 30 minutes, and is maximal after 1 hour at 4°C (Fig.4B). We typically fix the incubation time at 1 hour for enhanced sensitivity; however, this should be balanced with the rates of GTP hydrolysis occurring in the samples (see below).
Stability of the interaction
When bound to GST-PBD, active GTPase stays tightly bound and is not readily exchanged for another GTPase. Fig.5A shows that the binding occurring between Rac1-[35S]GTPS and GST-PBD is not reversed by addition of increasing amount of cold Rac1-GTPS, unless it is added at the same time as Rac1-[35S]GTPS.
We determined whether the binding of
GST-PAK1 PBD to active GTPase is sufficient to block GTP hydrolysis.
Recombinant Rac1 or Cdc42 was loaded with [32]P-GTP and
GTPase activity was measured at 20°C for 20 minutes.7
The binding of active Rac and Cdc42 to GST-PBD under these conditions
does not totally inhibit GTP hydrolysis, although the rate of
hydrolysis was significantly decreased by ~ 30%-50% (Fig.5B). GTP
hydrolysis is largely decreased at 4°C, so we recommend cell lysate
preparation and affinity precipitation should be performed at 4°C
or on ice, and all buffers used should be kept ice-cold.
Figure legends
Figure 1. Principle of the affinity precipitation assay to detect active Rac and Cdc42 using the GST-PAK1 PBD. (Note: x and Y represent non-relevant proteins in the lysate).
Figure 2. Analysis of GST-PAK1 PBD
preparation on a 12% SDS-PAGE. Lane 1: E.coli lysate, lane 2:
unbound proteins in E.coli lysate, lane 3: Sepharose
bead-isolated GST-PBD.
Figure 3. GST-PBD affinity precipitation assay performed with in vitro loaded GTPase.
A. Precipitation with GST-PBD using recombinant Rac1 either unloaded (-), loaded with GDP, or loaded with GTPS. The first lane represents the total Rac1 in the sample prior to PBD precipitation.
B. Precipitation with GST-PBD using BHK
cell lysates expressing various Rac1 mutants and blotted with an
anti-myc antibody to detect transfected GTPases only. Rac1 wildtype
(WT) was either precipitated as is (GDP-form) or after loading with
GTPS. Rac1 Q61L is constitutively GTP-bound, while Rac1 T17N is in the
GDP-bound state.
Figure 4.
A. Binding of increasing amount of GST-PBD to 20 pmoles of recombinant Rac1 loaded with [35S]GTPS (solid lines) or [3H]GDP (dashed lines).
B. Time course of the binding between
500 pmoles GST-PBD and recombinant Rac (50 pmoles) loaded with [35S]GTPS
(solid lines) or [3H]GDP (dashed lines).
Figure 5.
A. Stability of the PBD-GTPase interaction: 20 pmoles of Rac1-[35S]GTPS were incubated for 1 hour with (solid lines) or without (dashed lines) 4g GST-PBD before the addition of increasing amount of cold Rac1-GTPS and incubation for another hour. The radioactivity bound to GST-PBD was then quantified.
B. GTPase activity: 50 ng of Rac1 was loaded with [32P]GTP, then GTP hydrolysis was initiated at 20°C in the presence (solid lines) or the absence (dashed lines) of 4 g of GST-PBD.
References
1. P.D. Burbelo, D. Drechsel and A. Hall, J. Biol. Chem. 270, 29071 (1995).
2. G. Thompson, D. Owen, P. Chalk, and P.N. Lowe, Biochemistry 37, 7885 (1998)
3. V. Benard, B.P. Bohl and G.M. Bokoch, J. Biol. Chem. 274, 13198 (1999).
4. J-P. Mira, V. Benard, J. Groffen, L.C. Sanders and U.G. Knaus, Proc. Nat. Acad. Sci.USA 97, 185 (2000).
5. A.J. Self and A. Hall, Methods in Enzymol. 256, 67 (1995).
6. N. Geijsen, S. van Delf, J.A.M. Raaijmakers, J-W.J. Lammers, J.G. Collard, L. Koenderman and P.J. Coffer, Blood 94, 1121 (1999).
7. T. Akasaki, H. Koga and H. Sumimoto. J.
Biol. Chem. 274, 18055 (1999).