Protein Interaction Detection by Far-Western Blotting Protocol

Overview

Introduction of Far-Western Blotting

Far-western blotting is a method of characterizing protein–protein interactions, in which protein samples of interest are separated by gel electrophoresis, immobilized on a membrane, and then probed with a non-antibody protein. The term "far-western" was derived from western blotting, a similar method in which mem branes are probed directly with specific antibodies, and is also referred to as a west-western or blot overlay assay. Non-antibody proteins have been also used as a means to screen phage based expression libraries. Far-western blotting is very different from other commonly used methods to detect and characterize protein–protein interactions, and therefore complements these other approaches. Because the probe protein directly binds to denatured/separated proteins immobilized on a membrane, far-western blotting detects only direct interactions; by contrast, most non-far-western protein binding assays, such as immunoprecipitation and pull-down assays, may detect both direct association (two proteins make contact directly) and indirect association (two proteins do not make con tact, and another molecule in the ternary complex mediates the association). Thus, the far-western assay is often used to confirm direct interaction following immunoprecipitation or pull-down assays. The ability of far-western blotting to detect direct interactions is offset by limitations in the types of protein–protein interactions that can be detected. Because target proteins in a cell lysate are usually denatured in the process of gel electrophoresis, it may be difficult or impossible to detect interactions that require the native, folded conformation of the target protein. For this reason, far western blotting has been particularly useful in characterizing the binding partners of modular protein binding domains that bind to short, linear peptide motifs. It is now apparent that many signaling proteins interact with their partners via such modular binding domain–peptide interactions, and thus, the far-western approach is quite useful for analysis of signaling networks. However, these differences highlight the importance of using multiple approaches to assess specific protein–protein interactions. In far-western blotting, either a whole protein or fragment of a protein containing a suspected binding interface is used to probe interaction partners immobilized on a membrane. The interaction is visualized by direct labeling of the probe or by its subsequent detection with antibodies. There are a number of considerations in selecting the specific probe. First, ease of growth and purification of the probe must be considered. For the sake of cost and convenience, expressing probe proteins in bacteria is advantageous. However, only relatively small proteins (less than ~100 kDa) tend to remain soluble when grown in bacteria, so in general a fragment of a protein containing only the known or suspected binding domain will be easier to work with than the full-length protein. Second, it is useful to fuse the probe protein or domain to a tag sequence for ease of purification and detection. We routinely make probe proteins as glutathione S-transferase (GST) fusions, which has the dual advantage of allowing purification of proteins on glutathione columns, and allowing detection of bound probe with glutathione conjugates or with anti-GST antibodies. A further advantage of GST fusions is that GST exists as a stable dimer in solution. As in the case of antibodies, a dimeric probe binds with much greater avidity compared to a monomer to targets containing multiple binding sites, such as a membrane surface bearing many molecules of a target protein. In contrast to western blotting where a target protein is usually known in advance, far-western blotting can detect proteins on the basis of presence or absence of binding sites without any previous knowledge about their identities. From the intensity of bands observed on a far-western blot of a complex mixture of proteins, one can gain insight into both the number and relative affinity of binding partners for the probe in that sample. Furthermore, since some protein binding domains recognize their targets only after specific post-transcriptional modifications, far-western blot ting can be used to assess the modification status of multiple proteins in a sample. In this chapter, we will present a specific example of the utility of far-western blotting methods, in which GST-tagged Src homology 2 (SH2) domains, which bind specifically to tyrosine-phosphorylated target proteins, are used to probe the state of tyrosine phosphorylation of cellular proteins. Since quantitative comparison of interactions between multiple SH2s or other signaling proteins and their ligands is often of interest, we also provide a batch quantification method for multiple far-western blots.

Figure 1. Comparison of western and far-western blotting.

Materials of Far-Western Blotting

Subcloning of GST-SH2 Construct

pGEX-6P1.
Luria-Bertani (LB)-ampicillin agar plate: LB agar plate with 100 µg/ml ampicillin.
Phusion DNA polymerase.
Custom oligonucleotide primers.
Competent bacteria (strain NB42 or DH5α).

Evaluation of GST-SH2 Clones

LB-ampicillin: LB broth with 50 µg/mL ampicillin.
Isopropyl-β-D-thiogalactoside(IPTG).
Bacteria Triton X-lysis buffer (BXB): Phosphate buffered saline (PBS) with 100 mM ethylene diamine tetraacetic acid (EDTA), 1 % Triton X-100; add phenylmethyl sulphonyl fluoride (PMSF) to 1 mM, aprotinin to 1 % v/v (3 trypsin international units (TIU)/mL), dithiothreitol (DTT) to 1 mM just before use.
Sonicator with microtip probe.
5× sample buffer: 0.3 M Tris–HCl pH 6.8, 10 % sodium dodecyl sulfate (SDS), 25 % β-mercaptoethanol, 0.1 mg/mL bromophenol blue, 45 % glycerol.
Glutathione Sepharose 4B.
12 % SDS-polyacrylamide gel electrophoresis (PAGE) mini gel (see Subheading 2, item 4).
Control lysates (see Subheading 2, item 4).
Anti-phosphotyrosine antibody.
Coomassie blue solution: 40 % methanol, 10 % acetic acid,0.25 % Coomassie blue R-250.
Fixing solution: 20 % methanol, 10 % acetic acid.
Bacteria stock solution: 50 % glycerol (autoclaved).
Cryogenic tubes.

Large-scale Preparation of GST-SH2 Probe

Tris-NaCl-EDTA (TNE) buffer: 50 mM Tris–HCl 7.4, 150 mM NaCl, 10 mM EDTA; add aprotinin to 1 % (3 TIU/mL), PMSF to 1 mM just before use.
Chromatography column (poly-prep, 0.8 × 4 cm).
Elution buffer: 20 mM reduced glutathione, 100 mM Tris-HCl, pH 8.0; add aprotinin to 1 % (3 TIU/mL), PMSF to 1 mM just before use.
Sephadex G-25 PD-10 column.
Dialysis membrane tubing (molecular weight cut-off 3,500).
PBS with 10 % glycerol.
Bio-Rad Bradford dye reagent.
Ultrafiltration membrane.

Far-Western Blotting

Kinase lysis buffer (KLB): 150 mM NaCl, 25 mM Tris–HCl pH 7.4, 5 mM EDTA, 1 % Triton X-100, 10 % glycerol, 0.1 % sodium pyrophosphate, 10 mM β-glycerophosphate, 10 mM sodium fluoride (NaF); add aprotinin to 1 % (3 TIU/mL), PMSF to 1 mM, pervanadate (50 mM orthovanadate, 4 % hydrogen peroxide) to 50 µM just before use.
Sodium orthovanadate: Dissolve powdered sodium orthovanadate (final concentration will be 50 mM, but leave some extra volume for multiple rounds of pH adjustment); adjust with NaOH to pH of 10 (solution will turn bright yellow); boil in microwave until colorless, then stir until cooled to room temperature; adjust pH once again to 10.0, and repeat boiling; continue boiling and adjusting pH as above until pH stays at 10.0 after boiling (usually three rounds total); adjust volume for 50 mM, filter, and store at room temperature (RT).
Pervanadate solution: Mix 16 µL 30 % (w/w) hydrogen peroxide and 100 µL 50 mM sodium orthovanadate; incubate at room temperature for 30 min (not stable, needs to be freshly prepared before use).
Positive control lysates: Combine equal amounts of KLB lysate from pervanadate-treated NIH 3T3, HepG2, A431, and MR20 cells.
Negative control lysate: Prepare lysates of each cell line in the absence of vanadate, combined, and treated with tyrosine phosphatase PTP-1B for 1 h at RT.
12 % SDS-PAGE gel (1 mm thickness, for three mini gels).
Electrophoresis buffer: 25 mM Tris, 192 mM glycine, 0.1 % SDS.
Nitrocellulose membrane (0.2 µm pore size).
Transfer buffer: 10 mM 3-(cyclohexylamino)-1- propanesulfonic acid (CAPS), pH 11.0, 20 % methanol, kept at 4 °C.
Tris buffered saline-Tween 20 (TBST): 25 mM Tris–HCl, pH 8.0,150 mM NaCl, and 0.05 % (v/v) Tween-20.
Blocking solution: 10 % fat-free milk in TBST, 1 mM EDTA, 1 mM sodium orthovanadate.
Labeling reagent: Glutathione-HRP conjugate or anti-GST HRP conjugate.
Chemiluminescence kit.
Imaging: X-ray film; CCD detection system.

Stripping and Reprobing

Acidic stripping buffer: 100 mM Glycine-HCl pH 2.0.
SDS-ME stripping buffer: 2 % SDS, 100 mM β-Mercaptoethanol,62.5 mM Tris–HCl pH 6.8.

Probing of Replica Membranes

Same as Subheadings Far-Western Blotting and Stripping and Reprobing.

Image Adjustment

Photoshop CS.

Batch Quantification

ImageJ
Probe Reader plug-in.

Methods of Far-Western Blotting

As for western blotting, protein samples are separated by SDS- PAGE and transferred to a nitrocellulose or polyvinylidene fluoride (PVDF) membrane. The blocked membrane is then incubated with a probe followed by appropriate wash, and bound probes are visualized. Generally any protein samples compatible with western blotting can be used including whole cell lysates, purified proteins, and native or denatured samples. The far-western method described here is a rapid and simple protocol in which the membrane is prepared without denature–renature procedures, the probe is labeled directly, and probing is performed in one step. This protocol has been optimized to detect in vitro interaction between modular binding domain probes and immobilized proteins containing short peptide motifs (see Note 1). Of note, alternative protocols are available including another example in which proteins containing modular domains on a membrane are probed with labeled binding motifs. Below, we present a specific protocol for generating GST-SH2 domain probes and using them to probe tyrosine phosphorylated whole cell lysates. We have also included a method for aligning and quantifying multiple far-western blots which allows for comparative assessment of SH2 binding. Of course these procedures can be adapted for any modular protein binding domain and its binding partners with minor modifications. For all far-western blotting methods, detection of specific signal is strongly dependent on the quality of the probe protein. Insolubility, aggregation, or denaturation of the protein tends to cause nonspecific background, and even the native probe may bind nonspecifically to abundant proteins in the sample. Thus, it is important to: (1) confirm purified probe is soluble, folded, and not significantly degraded; (2) evaluate activity of the probe and optimize binding conditions if needed; and (3) always include appropriate positive and negative controls for each experiment to ensure any positive signal is indeed specific. To address these considerations, in the following section we will describe a detailed protocol for generation and evaluation of GST-SH2 fusion probes. Appropriate controls should be prepared considering the intended physiological activity of the probe. At minimum, GST alone, or more ideally GST fused to the domain of interest bearing a mutation known or suspected to abolish specific binding activity, should be used as a negative control probe.

Subcloning of GST-SH2 Construct

Retrieve cDNA and protein sequences of a SH2 domain containing protein of interest, e.g., at NCBI Entrez Gene(see Note 2).
Find location of the SH2 domain using the protein sequence at ScanProsite.
Find the nucleotide sequence corresponding to the SH2 domain using the sequence editor program (see Note 3).
Find academic or industry source for corresponding cDNA (see Note 4), otherwise clone the cDNA by RT-PCR method.
Design primers for PCR (see Note 5).
Amplify the SH2 fragment by PCR using the oligonucleotide primers and the cDNA template.
Digest the PCR product, and then purify the fragment from an agarose gel.
Insert the purified SH2 fragment into a pGEX vector digested with appropriate restriction enzymes.
Transform competent bacteria and grow overnight on LB ampicillin agar plate (see Note 6).

Evaluation of GST-SH2 Clones

Protein expression and solubility of GST-SH2 clones can be tested quickly in small-scale bacterial cultures (see Note 7), and the activity of the probe can be tested at the same time by a control pulldown assay.

Inoculate 0.4 mL dense liquid culture of bacteria into 1.6 mL fresh pre-warmed LB-ampicillin.
Shake at 37 °C for 1 h.
Add IPTG to 0.1 mM and shake at 37 °C for 1 h to induce protein expression.
Transfer 1.5 mL bacteria to microcentrifuge tubes and spin at 10,000 × g at 4 °C for 2 min in microfuge.
Remove supernatant and resuspend bacterial pellet in 0.4 mL BXB.
Vortex to resuspend, then sonicate briefly (e.g., 2–5 s at relatively low power) on ice to break cells, let sit on ice, then repeat. Try to avoid foaming; if foaming occurs, let rest on ice for a few minutes to allow foam to dissipate.
Remove 10 µL of total lysate, add 2.5 µL 5× sample buffer for gel.
Spin rest of lysate for 5 min in microfuge at 10,000 × g at 4 °C, transfer supernatant to a new tube.
Take 10 µL of the cleared lysate for gel as above.
Take 100 µL of cleared lysate and add to 10 µL glutathione agarose bead slurry (cut end off pipet tip with razor blade to more accurately pipet beads).
Rotate at 4 °C for 30 min.
Spin out briefly, and wash beads 3× with 1 mL cold BXB.
Resuspend the bead pellet with BXB, take 10 µL, and add 2.5 µL 5× sample buffer for gel (see Note 8).
Boil all samples and run on 12 % SDS gel.
When gel is done, stain for 15 min with Coomassie blue solution, de-stain with fifi x solution, and then dry(see Note 9).

Steps 16-19: Evaluate pTyr binding activity by GST pull-down assay (optional).

Incubate remaining beads (GSH-bound fractions for pGEX-SH2 clones and a control pGEX clone, if any) with 10 µg cell lysates (see Note 10).
Rotate for 1 h at 4 °C, and wash three times with BXB.
Boil all samples and run on 12 % SDS gel (see Note 11).
Perform western blotting with anti-phosphotyrosine antibody(see Note 12).
Subject positive clones to DNA sequencing and store bacteria in 25 % (v/v) sterile glycerol at –70 °C.

Large-Scale Preparation of GST-SH2 Probe

GST-SH2 probe is purified following the standard protocol for preparation of GST fusion proteins using pGEX series bacterial expression vectors (see Note 13).

Inoculate frozen stock culture of a verified GST-SH2 clone in 50 mL LB-ampicillin overnight.
Inoculate 50 mL dense overnight culture to 1 L LB- ampicillin. Shake at 37 °C for 2 h (see Note 14).
Add IPTG to 0.1 mM. Shake at 37 °C for 3 h (see Note 15).
Centrifuge at 5,000 × g at 4 °C for 10 min.
Resuspend pellet in 5–20 mL ice-cold BXB, transfer to 50 mL tube, and sonicate on ice until cells are broken (see Note 16).
Centrifuge at 5,000 × g at 4 °C for 10 min to remove debris (see Note 17).
Add glutathione-agarose to supernatant: 3 mL 50 % (v/v) bead slurry/L original culture.
Rotate at 4 °C about 1 h (up to 2 h).
Wash beads with TNE buffer: spin out beads at low speed, remove supernatant, and resuspend beads in fresh buffer. Repeat 3–5×.
To elute GST-SH2, pour beads into small disposable column, elute with approximately 3 bead volumes of elution buffer (see Note 18).
Change buffer by gel filtration on a Sephadex G-25 PD-10 column according to the supplier's instructions. Briefly, equilibrate column with approximately 25 mL PBS-10 % glycerol. Discard the flow-through. Add sample followed by buffer up to a total volume of 2.5mL. Discard the flow-through. Elute with 3.5 mL buffer (collect seven 0.5 mL fractions of the eluate in separate tubes (see Note 19).
Estimate relative protein concentration with Bio-Rad Bradford dye reagent, combine top three fractions into one tube (see Note 20).
Determine protein concentration by Bradford assay, take 500 µg of protein and dilute to 0.1 µg/µL with PBS-10 % glycerol, and aliquot 50 µL diluted probe into chilled micro centrifuge tubes. Store the aliquots and undiluted probes at –70 °C (see Note 21).
Evaluate the purification fractions by 12 % SDS gel as in previous Subheading evaluation of GST-SH2 clones (see Note 22).

Far-Western Blotting

Separate proteins on SDS-polyacrylamide gels and transfer to nitrocellulose or PVDF membranes following general western blotting protocol (see Note 23).
Block membranes in blocking solution for about 1 h at room temperature or at 4 °C overnight (see Notes 24 and 25).
To label probe, thaw the stored probe on ice and add 5 µL GSH-HRP conjugate (0.1 µg/µL) to 50 µL of diluted probe (0.1 µg/µL) (see Note 26).
Incubate on ice for about 1 h. Labeled probes can be stored at 4 °C (see Note 27).
Dilute labeled probe to optimal concentration with blocking buffer and apply to the blocked membrane (see Note 28).
Let probe bind for 1–2 h at room temperature, then wash with multiple changes TBST for 20 min.
Visualize signal by enhanced chemiluminescence according to manufacturer's instruction (see Note 29).
Take appropriate exposure using x-ray films or an image analyzer (see Note 30).

Stripping and Reprobing

Generally, fresh membranes are best for far-western blotting; stripping and reprobing of the membrane may result in significant signal loss and increased non-specific background. Nevertheless recycling membranes might be beneficial if sample is limited, or if precise comparison of specific bands is needed within the same membrane.

After initial probing, keep membrane wet; it can be stored wet, wrapped in plastic wrap, at 4 °C (up to a week) or −20 °C (for longer period).
Rinse the membrane twice with TBST.
Immerse membrane in stripping buffer at room temperature for 20 min under gentle rocking agitation (see Note 31).
Wash membrane in large volume of TBST at room temperature for 45 min with frequent buffer changes.
Proceed to blocking and reprobing.

Probing of Replica Membranes

Although a single far-western blot is useful for identifying differences between samples, comparison of probe specificities, such as the phosphosite preference of different SH2 domains, requires parallel probing and quantification of multiple replica membranes. Here we describe a method for the preparation, probing and detection, image adjustment, and batch quantification of multiple membranes.

Prepare multiple nitrocellulose or PVDF membranes of equal size.
Draw a frame position mark along the outside edge of each membrane using a permanent marking pen (see Note 32).
Run aliquots of lysate samples on multiple protein gels and transfer using identical experimental conditions.
For each membrane, follow steps 2-8 of Subheading Far-Western Blotting Keep probing and image acquisition conditions such as blocking, probing, washing stringency, and imager machine settings as constant as possible for all membranes.
Following far-western image acquisition, acquire a reference image of the blot under room light before removing it from the imager. Be sure that the frame position mark is clearly visible (see Note 33).
If necessary, strip and reprobe membranes according to steps 1-5 of Subheading stripping and reprobing.
After all far-western blotting has been completed, perform western blotting with an anti-phosphotyrosine antibody for all membranes.
Visualize signal by enhanced chemiluminescence with appropriate exposures using the same imaging method used for far-western probing (see Note 34).
Acquire a white light reference shot of the blot before removing it from the imager.

Image Adjustment

For accurate batch quantification, all images must be well aligned. Images from the same physical membrane, e.g., reprobing, can be easily aligned using the frame position marks, while those from different membranes must be carefully aligned using the anti-phosphotyrosine or other reference blot images.

Export imager fifi les in TIFF or high resolution JPEG format, and open as a multi-layer image with image editor software such as Adobe Photoshop.
Align blots derived from the same membrane using the room light images acquired in step 5 of Subheading probing of replica membranes. Utilize the frame position marks to align and overlay membranes using the rotation and scaling functions.
For images from different membranes, align the anti-phosphotyrosine bands by utilizing editing functions such as rotation, scaling, and distortion to correct minor inconsistencies in lane and band shape (see Note 35).

Batch Quantification

Bin-based batch quantification partitions each far-western lane into a set number of equal sized rectangular regions of interest using a grid. This method allows for an unbiased quantification of far-western blots by capturing the differences in SH2 binding patterns produced by each probe and compensating for minor residual variations in band position. Since all blots are quantified as a batch, this method also ensures every far-western is partitioned equally and quantified using the same parameters. While there are multiple possible ways to partition far- western data for quantification (i.e., molecular weight standard or band-based) we have found that a bin-based quantification works well when gels/blots are relatively homogenous in shape and have many bands that need to be quantified.

Create a mock-up table with approximately 20 rows and a column for every sample lane using Excel. Paste the table on top of the aligned blot image layers in Photoshop. Scale the table so that the columns are aligned with sample lanes (see Note 36).
Export compiled images, including the grid, as ImageJ compatible files.
Import all aligned images to ImageJ, or comparable image quantification tools, as an image sequence (see Note 37).
Use the ImageJ default background subtraction tool, invert, and select the Probe Reader plug-in.
Create a quantification grid with dimensions (size, bin number and arrangement) equal to that of your mock-up grid.
Press the green square in the upper left corner of the window to perform batch quantification. The data will be saved to the ImageJ application directory as a single tab-delimited file containing a matrix of bin quantifications.
Open the data fi le with Excel or another spreadsheet software package. Data in the matrix are the background subtracted raw values (see Note 38).
Scale or normalize the data to the image specific maximum bin value or other values as appropriate for comparable band analysis.

Notes of Far-Western Blotting

This approach, termed a reverse-phase assay, maximizes signal to noise when the concentration of analyte is low (e.g., tyrosine- phosphorylated proteins in whole-cell lysates), because effificient binding to the immobilized analyte can be driven by high concentrations of the SH2 domain probe in solution.
Useful information about modular interaction domains is available at web-based databases:

Addgene.
NCBI Entrez Gene.
ScanProsite.
The Human Protein Reference Database.
UniGene.
SH2 domain information site.

Several sequence editing programs are available, e.g., CLC SEQUENCE VIEWER.
Large collections of full-length or partial cDNAs are now commercially available at reasonable prices, e.g., IMAGE Consortium. To find appropriate cDNA clone, go back and forth between NCBI Entrez Gene site and Unigene sites; referring to the SH2 nucleotides, find IMAGE clones that contain intact SH2 domain. Mayer/Machida Lab GST-SH2 plasmids (complete set of human SH2 domains) can be obtained from Addegene.
To ensure maximal binding activity and solubility, we usually include 5–10 amino acids N- and C-terminal to the domain boundary for an SH2 domain probe. In some cases, yield and solubility of the fusion protein are greatly affected by the precise borders of the construct. Structural studies, if any, are useful when planning the borders for any modular domain. PCR primers should have approximately 24 nucleotides of homology to the template; the 5′ end should have five C's (this allows efficient digestion with restriction nucleases) followed by a restriction site to be used for cloning in-frame into the pGEX expression vector, followed by the region with homology to template.
We routinely use E. coli NB42, which lacks the two major proteases; this may help increase yields by limiting degradation. Similar results can be obtained with typical laboratory strains such as DH5α, though yields may be somewhat lower and bacterial growth is slower. We do not routinely perform restriction endonuclease mapping or DNA sequencing of plasmids at this step, instead we immediately test protein expression of clones as below.
Solubility is one of the major determining factors of probe activity in far-western blotting. In our hands about half of the GST-SH2 domains are relatively insoluble, and highly insoluble domains give poor yields and tend to lack detectable binding activity. Multiple strategies to improve solubility have been reported. We have observed that, for about two-thirds of insoluble GST-SH2 proteins, solubility could be significantly improved when protein is expressed in bacteria at lower temperature.
If solubility of the GST fusion is unknown, volume used to resuspend beads should be adjusted to load two- to tenfold more of the bead-bound fraction relative to total cell fractions on gel to visualize the band.
A GST-SH2 band of about 40 kDa in size, although the degree of protein expression and solubility may vary depending on the constructs. If protein degradation is observed on gel, care must be taken in the large-scale purification.
Use positive and negative control lysates for pull-down; as an active SH2 domain should bind to tyrosine-phosphorylated proteins, we use pooled lysates of pervanadate-treated cell lines as a positive control (pervanadate inhibits endogenous protein tyrosine phosphatases, thus strongly enhancing tyrosine phosphorylation in vivo). Corresponding pooled lysates lacking tyrosine phosphorylation (prepared in the absence of phosphatase inhibitors and then treated with phosphatase in vitro) serve as a negative control.
The gel lanes should contain: positive control lysate 5 µg; negative control lysate 5 µg; pull-down positive control with GST bound beads; pull-down negative control with GST-bound beads; pull-down positive control with GST-SH2-bound beads; pull-down negative control with GST-SH2-bound beads.
In the pull-down result, a functional GST-SH2 protein should have increased anti-pTyr signal relative to GST control. If GST-SH2 is correct size (and sequence) but the pull-down result is not clearly positive, activity of the probe could be reevaluated by far-western assay.
We routinely obtain about 5–10 mg fusion protein per liter of bacteria using the protocol described here (less if the protein is less soluble).
Usually an OD of 0.4–0.6 is optimal at this step.
If the protein is highly insoluble, consider protein expression at lower temperature: cool down culture with ice before adding IPTG, then add IPTG to 0.1 mM and shake at 30 °C for 4 h or 15 °C about 16 h.
Sonication time 2–3 min total with microtip at power setting 3–4; solution should become slightly darker and less turbid looking. Avoid foaming and overheating the lysate (let solution rest on ice if it warms up detectably).
Supernatant does not have to be absolutely clear at this step.
If the elution yield is found to be suboptimal, the following conditions may help: shorter incubation time of lysate with glutathione beads (~30 min); higher pH (up to 9.6) and glutathione concentration (up to 50 mM) in elution buffer; iterative batch elution (2–3 × 30 min).
Alternatively, the eluate can be dialyzed overnight against several large volumes of PBS-10 % glycerol.
To quickly check relative protein concentration, add 2 µL of each PD-10 elution fraction to 50 µL diluted Bradford dye and vortex. Pool the top three fractions with brightest blue color. If color change is not obvious (this occurs when protein con centration is less than 0.2 µg/µL), take third to fifth fractions and proceed to ultrafiltration for concentration (below).
If protein concentration is low (<0.5 µg/µL or so), sample can be concentrated by ultrafiltration according to manufacturer's instructions.
It is important to take aliquots at every step of purification. Take the same fraction of the total at each step (e.g., 1/5,000 of total prep) to monitor percent recoveries at each step. Retain the pellet after sonication and the supernatant after bead binding at –70 °C in case the majority of the fusion protein is there. If final product is highly degraded, we recommended revaluating parameters of purification such as prep scale, temperature of induction, protease inhibitors, EDTA concentration, and sonication strength.
Amount of protein loading on gel is important for an optimal result, and thus should be carefully considered. We usually load 10–50 µg of whole-cell protein lysate per lane for far-western blotting with GST-SH2 domains. We routinely use a nitrocellulose membrane with 0.2 μm pore size. We find the advantages of nitrocellulose compared to PVDF are ease of use (no pretreatment is needed) and better signal-to-noise ratio (our observation).
Composition of blocking solution is important to maximize signal to noise. Strong blocking buffer lowers background but may decrease or eliminate specific signal. We obtained best results with 10 % nonfat milk-based buffer for GST-SH2 domain probes. For other probes, optimal conditions should be determined empirically; blocking buffer containing 5–10 % non-fat milk (strongest), 1–5 % bovine serum albumin (BSA) (moderate), or 1–5 % ovalbumin (moderate) can be tested.
Transferred membrane can be stored at −20 °C for later use: rinse with TBST once and wrap in plastic wrap. Care should be taken to avoid damaging the membrane, which causes undesirable background signal.
GSH-HRP (1 µg/µL in PBS-10 % glycerol) can be stored at −70 °C. Recently Sigma has discontinued production of this conjugate. If GSH conjugate is unavailable, we have found that anti-GST-HRP conjugate can be used in the same way, with some decrease in signal strength. To label GST fusion probe, add 0.2 µL (~3 µg) antibody to 50 µL (5 µg) of diluted probe (0.1 µg/µL). This direct labeling method presumably promotes oligomerization of probe and HRP, increasing the avidity of binding to the target and enhancing signal. In addition, this one-step probing and washing procedure is time saving and may avoid dissociation of probe during washes if probe binding is weak.
Purification of the labeled probe is not needed. Labeled probe can be stored at 4 °C with protection from light up to a month.
Optimal concentration (maximum signal and low nonspecific background) should be determined empirically in pilot experiments. For GST-SH2 domains, optimal concentration ranges from 0.01 to 0.15 µM (most typically 0.05 µM). Probing can be done either in a sealed nylon bag or in an open tray with a sufficient volume of solution (1–10 mL/10 cm 2 membrane) with occasional agitation. Do not allow the membrane to dry at any point of the probing procedure through detection.
Select chemiluminescence kit with appropriate sensitivity. It is important to cover whole blot area evenly during incubation, and to drain off excess chemiluminescence reagent well (retained solution causes high background).
Like western blotting, multiple parameters of the assay influence the level of signal and non-specific background, such as blocking buffer, probe concentration, incubation time and temperature, washing stringency, and detection system. Thus, optimal conditions for a given GST-SH2 or other GST fusion probe should be customized empirically. In general, weak signal can be enhanced by decreased concentration of non-fat dry milk in blocking buffer, increased probe concentration, milder washing, more sensitive chemiluminescence kit, and longer exposure times. Alternatively, domain concatenation may help.
If stripping is not sufficient with this method, try a more stringent condition using SDS-ME stripping buffer: briefly, immerse membrane at 50 °C for 30 min under vigorous agitation, and then wash with TBST for 60 min with frequent shaking.
For quantitative far-western blotting, it is best to obtain blot images using a digital imager. Having image fi les of your blots will significantly improve the ease of data analysis.
The reference markings on the room light image will be used as a reference for aligning multiple images obtained using a single membrane.
Anti-phosphotyrosine blots will be used as a reference to align and adjust far-westerns images. Depending on the bait protein, different types of reference blots including general protein staining should be considered.
For accurate comparison of SH2 domain binding patterns in different blots, gel-to-gel variations due to bended lanes and bands, smiles, bubbles, nonuniform background, etc, should be diminished experimentally or computationally. In terms of experimental reproducibility, we routinely use NuPAGE Bis Tris Precast Gels for their higher protein band resolution and relatively minor gel distortion.
Although the number of rows and columns is user customizable, empirically a 20 row grid is sufficient to distinguish bands with slightly different molecular weights on a 4–12 % mini gel. The number of rows can be determined by comparing quanti-fification data from replicated blots, as this should result in nearly identical band profiles if optimal.
For ImageJ, we created a plug-in (probe_reader2.jar) for batch quantification of multiple Western/far-Western blots. It is best to name each image file sequentially using numbers, as ImageJ and the Probe Reader will sort images in numerical order based on file name.
Data in each cell should correspond to the bin position on your grid. Bin quantifications from each image are stacked vertically on the spreadsheet according to their image sequence order.