Western or Enzyme-Assisted Imunoelectro-Blotting (IEB)



Ed Rybicki and Maud Purves

Copyright Ed Rybicki, 1996





Electroblotting has been a feature of a large number of laboratories for a (relatively) long time now, and there are a large number of different apparati around that will efficiently transfer proteins (or other macromolecules) transversely from gel to membrane. Most of these, however, are based on the design of Towbin et al. (1979): ie., they have vertical carbon / stainless steel / platinum electrodes in a large tank .

Some enterprising newcomers have fairly recently turned the whole issue on its side, and have started "semi-dry" or "horizontal" blotting. With this technique, one uses two plate electrodes (stainless steel or graphite / carbon) for uniform electrical field over a short distance, and sandwiches between these up to six gel / membrane / filter paper assemblies , all well soaked in transfer buffer. The assembly is clamped or otherwise secured ON ITS SIDE, and electrophoretic transfer effected in this position, using as transfer buffer only the liquid contained in the gel and filter papers or other pads in the assembly.

There are a number of advantages to this procedure over the conventional upright protocol, not the least of which is that as little as a couple of hundred millilitres of buffer are all that is needed for electroblotting everal gels, compared to as much as five litres for some commercial kits. In addition, several gels can be blotted simultaneously; electrodes can be cheap carbon blocks; less power is required for transfer (and therefore a simpler powerpack).


The reason for transferring proteins to membranes from gels is so as to be able to get at them more efficiently with various probes, as polyacrylamide is not particularly amenable to the diffusion of large molecules. The most popular type of probe of immobilised proteins is an antibody of one type or another: the attachment of specific antibodies to specific immobilised antigens can be readily visualised by indirect enzyme immunoassay techniques, usually using a chromogenic substrate which produces an insoluble product. Lately chemiluminescent substrates have begun to be used because of their greater detection sensitivity. Other possibilities for probing include the use of fluorescent or radioisotope labels (fluorescein, 125I). Probes for the detection of antibody binding can be conjugated anti-immunoglobulins (eg: goat-anti-rabbit / human); conjugated staphylococcal Protein A (which binds IgG of various species of animal); or probes to biotinylated / digoxigeninylated primary antibodies (eg: conjugated avidin / streptavidin / antibody).

The immunoassay is normally done by blocking the transfer membrane with a concentrated protein solution (eg: 10% foetal calf serum, 5% non-fat milk powder) to prevent further non-specific binding of proteins; this is followed by incubation of the membrane in a diluted antiserum / antibody solution , washing of the membrane, incubation in diluted conjugated probe antibody or other detecting reagent, further washing, and the colorimetric / autoradiographic / chemiluminescent detection.

The power of the technique lies in the simultaneous detection of a specific protein by means of its antigenicity, and its molecular mass : proteins are first separated by mass in the SDS-PAGE, then specifically detected in the immunoassay step.

It is also possible to use a similar technique to elute specific antibodies from specific proteins resolved out of a complex mixture , many of whose components react with a given antiserum: one can electrophorese a mixture of proteins, cut out a specific band from a gel or membrane, and use this to fish out specific antibodies from a serum.

Staining of proteins in gels

Staining of proteins in gels may be done using the standard Coomassie brilliant blue (or PAGE blue), Amido Black, and more lately, silver stain reagents of different kinds. All of these protocols are relatively long: silver staining requires a number of finicky treatment and washing steps and takes at least a couple of hours, while other stains are two-step stain/destain procedures , but require hours (sometimes days) for satisfactory destaining. Silver staining is (or can be) extremely sensitive - to +1 ng/band) - while of the other commonly used stains, Coomassie brilliant blue G-250 is probably the best, but detects only to about 0.3 ug/band . None of these stains are easily reversible, and most often the protein cannot be recovered intact for other procedures.

However, it is possible to reversibly stain gels prior to blotting by a couple of methods: the simplest is by simply soaking them in ice-cold 1M potassium chloride : SDS precipitates as KDS, and proteins are visible as whiter zones in an opaque-to- translucent white background. The method is not sensitive, however.

Now Lee et al. (C. Lee, A. Levin and D. Branton: Copper staining: a five- minute protein stain for sodium dodecyl sulfate- polyacrylamide gels; Anal. Biochem. 166, 303-312; 1987) have described an alternative to common protein stains - and to KCl treatment - which is sensitive, stable, and totally reversible for subsequent recovery of the protein.


This protocol describes the use of horizontal blotting of simple SDS-PA gels, and subsequent detection of proteins using rabbit antisera and alkaline phosphatase-conjugated goat-anti-rabbit IgG, detected using bromo-chloro-indolyl phosphate (BCIP) and Nitro-blue tetrazolium (NBT) salts.

Apparatus and Equipment:

The Hoefer SE 600 vertical slab gel apparatus (or other) is used for SDS- PAGE.

Nitrocellulose paper (0.45 um pore size) is supplied as 33 x 300 cm rolls; filter paper is Whatman 3MM or - preferably - BLOTTING PAPER . Both are cut to size USING A CLEAN BLADE AND GLOVES!!!

Blotting Electrodes are carbon slabs, 1cm thick x 20 cm long x 15 cm wide, with heavy insulated leads, OR patent blotting apparatus.

Powerpack: normally a high-amperage low-voltage apparatus; the Shadon Southern Destainer powerpack is suitable. It must be capable of supplying 500mA - 1A current at voltages as low as 5V.

Blotting Buffer:

This is often simply Laemmli electrophoresis buffer (50mM Tris-HCl / 0.196M glycine pH 8.3) with 20% methanol added; however, this has too high an ionic strength for horizontal blotting because of the short inter-electrode distances used. We have found that 25mM Tris-Cl pH 8.3/20% methanol (or pH 7.0 - 8.8) works well; for that matter, one may equally well use 25 - 50mM phosphate/methanol pH 7.0 - 8.0 .

It is a good idea to use pH values above 7.0, in order to ensure that all proteins migrate towards the anode. The methanol is necessary to prevent gel swelling with heating, and to keep proteins adsorbed to the membrane.

My friend Jorge Meyer (then at CIAT, Colombia) wrote that one may also use anodal and cathodal buffers as follows: Anodal 1 (for pad nearest anode): 0.3M Tris/20%MeOH pH 10.4. Anodal 2 (for other gel pad assemblies): 25mM Tris/20% MeOH pH 10.4. Cathodal (for all pads etc. nearest cathode): 25mM Tris/40mM aminocaproic acid/20% MeOH, pH 9.4.

It turns out that blotting big gels (eg 14.5 x 11 cm) is quite adequately done (even for two or more stacked gels) at 0.3 A for 60 min or less: the system does not heat up appreciably, and transfers are good.

NOTE: you should cut nappy liners / papers to the size of the gel(s) to be blotted, so as to minimise surface area exposed to the electrode: this reduces the amount of current that needs to be passed in order to effect transfer (and incidentally reduces heating effects).


Freshly-electrophoresed SDS-polyacrylamide gels are dipped into transfer buffer (0.025 M Tris-HCl/20% (v/v) methanol, pH 8.3), then laid flat on pre- wetted nitrocellulose paper supported on three layers of transfer buffer- wetted filter paper resting on the ANODE (+ve electrode). We have previously made use of nappy liners as absorptive buffer-containing pads; however, blotting paper or (if really necessary) 3MM paper may be used instead. The gel is overlaid with three wetted filter papers, and then either the CATHODE (-ve electrode) or another layer of nitrocellulose / gel / blotting paper. Care should be taken to exclude bubbles between gel and nitrocellulose, and between nitrocellulose and paper.

The assembly is placed in a plastic tray resting on the anode. The ANODE (electrode on the nitrocellulose paper side of the assembly) is connected to the anode or RED connector - and the gel side CATHODE to the cathode or BLACK connector - of an appropriate powerpack. A current of 500mA is passed for 20-30 min to effect transfer.


Transfer is found to be essentially quantitative - for thin assemblies - after electrophoresis under these conditions. One hour should be sufficient for transfer of all but the most recalcitrant proteins. If in doubt, blot gels in duplicate, remove one and stain for protein after 1 hr and continue blotting other.


Disassemble assembly, saving nappy liner pads if used (put straight back into buffer on tray if to be re-used soon). Rinse membrane in saline or other buffer before further treatment.


Staining Proteins on Membranes:

Nitrocellulose or other adsorptive membrane is generally used for transfers: nitrocellulose is OK for adsorption and gives low staining backgrounds with India ink, Amido Black, Coomassie Brilliant Blue, colloidal gold and peroxidase and alkaline phosphatase substrates but is fragile; Amersham make a tough NC which may be more suitable. They also make Hybond-N (nylon) which is amazingly strong but used to give horrible background with enzyme-immune substrates; the Boehringer non-radioactive nucleic acid detection kit blocking agent gets around this. No nylon can be stained with the other stains; however, Millipore claims their polyvinylidene difluoride (PVDF) membranes are as good as nylon and can be stained with colloidal gold.

Methods which have been tried here are Amido Black staining (not satisfactory); India ink staining (OK), and Ponceau S staining (highly useful).

Two other techniques for detection on membranes:

Indirect enzyme immunoassay

Buffers and Solutions:

Incubation/Blocking Buffer: 10 mM Tris-Cl/150 mM NaCl containing 1-5% Protea non- fat milk powder and 0.05% Tween-20 or Triton X-100 (=Nonidet P-40), pH 7.4.

Washing buffer: 10 mM Tris-Cl/150 mM NaCl pH 7.4 ( OR SIMPLY SALINE ) containing o.05% Tween-20 or Triton X-100.

Substrate buffer: 100 mM Tris-Cl/100 mM NaCl/5 mM MgCl2 pH 9.5. Substrate stocks: Nitro Blue tetrazolium (NBT) (Sigma), 75 mg/ml in 70% dimethyl formamide; 5-Bromo-4-chloro-3-indolyl phosphate (BCIP) (Sigma), 50 mg/ml in formamide (100%).



Blocking: nitrocellulose blots are briefly rinsed in transfer buffer, then soaked for 1 hr at 37oC, or 2 hr at room temperature in blocking buffer. This procedure allows saturation of all non-specific protein binding sites on the blots.

Attachment of specific antibodies to proteins: is achieved by incubation of blots in rabbit antisera diluted 1/10- 1/1000 in incubation buffer, in sealed boxes for 1 hr on a shaking waterbath at room temperature (22oC ).

Washing: blots are washed by shaking in +100 ml/wash, 3x5 min., at room temperature. Increase number and duration of washes if background is a problem.

NB: Recognition of host plant (or other) proteins can be avoided

Probing of antibody binding: is by means of suitably (eg 1/5000)-diluted alkaline phosphatase - goat anti-rabbit globulins (GAR-AP)(Miles Laboratories, Cape Town) in incubation buffer. Incubation conditions and washing are the same as for rabbit antiserum.

Antibody binding to blots is visualised: by the reaction of NBT and BCIP mixture.

To make 10 ml of substrate: NB: MAKE FRESH BEFORE EACH ASSAY.

Mix 50 ul of NBT stock with 10 ml of substrate buffer, then add 50 ul of BCIP stock and mix well by swirling . NOTE: IF SOLUTION IS FLOCCULENT OR VERY CLOUDY AT THIS STAGE, DISCARD AND MAKE FRESH STOCKS . A little cloudiness is normal. Pre-rinse blot(s) in a little substrate buffer, then add substrate: +10 ml/100 cm2 blot. Incubate in the dark with occasional agitation. The reaction can take several hours, but is normally complete in 30-60 min. Leave overnight with gentle agitation IN THE DARK if reactions are very faint. Terminate reaction by washing with water. Blots are dried under weighted filter paper, and stored in the dark. Results may be recorded photographically using a red filter for B& W, or with no filter for colour slides.

The following Ponceau S method was culled from the Net: “Sambrook et. al on page 18.67 (Vol 3) outline a method using Ponceau S (which can be obtained from Sigma). The formula is a follows: 2 g Ponceau S 30 g trichloroacetic acid 30 g sulfosalicyclic acid water to 100 ml. I highly recommend beginning with less than 20 ml water as the two acids bring with embibed water. The stain works great and reverses with a normal TS (Tris/saline) wash.”

Philip Dykema

Two other techniques for studying protein- protein interactions on top of crosslinking and immunoprecipitation.

Far western blot: Similar to a western blot except that a labelled protein is used as a probe [ to detect specific protein-protein interactions ]. The probe can be biotinylated or labelled with 32P, 125I or 35S. The following references will be useful: Oncogene (1990), 5, 451-458 Mol Endocrinol (1991), 5, 256-266 Genes & Dev (1992), 6, 439-453 Science (1992), 257, 803-805

Affinity chromatography: Express one of your protein as a GST (or MBP) fusion in E. coli. These fusion proteins will then bind to gluthatione sepharose or amylose resins and can be used to fish out binding proteins. Numerous labs have used this technique which was originally described by Kaelin et al. in Cell (1991), 64, 521-532.”


NB: Recognition of host plant (or other) proteins can be avoided by diluting the antiserum in buffer containing (for example) a 1:3 mix of sap of healthy plants (plants crushed 1:1 (w/v) in 0.1 M phosphate, pH 7.0) or other antigen with the buffer used to dilute the antiserum; and pre-incubating at 37oC for 1 hour. This procedure gets rid of host plant and E. coli contaminant reactions. Another procedure more generally used in this laboratory is the removal of anti-host antibodies by plant or bacterial extracts immobilised on nitrocellulose (or other membrane): simply soak membrane in a concentrated extract of whatever it is that you want to remove antibodies with +30 min); block as for normal blot; add diluted antibody suspension to this membrane for +30 min before use on the Western blot proper, then pour off for use as probe antibody. This procedure should remove all antibodies responsible for reacting with antigens that contaminated the inoculum used to immunise rabbits (or whatever).  SEE ALSO HERE FOR MORE DETAILED PROTOCOLS!!!

Another option - got by email through the bionet.methods-reagents newsgroup - is the following: “We used to use the following as substrates for Western blotting. Yields a bright pinkish band which quite pleasing to the eye and slide- making but may be not that high contrast for the photography. 100 mg Fast Red TR salt (Sigma: f8764) 50 mg Napthol AS-MX phosphate (Sigma N5000). Each of these are dissolved in 50 ml of 0.5M tris pH 8.0, mixed in the tray and incubated with the membrane. Worked well for me.”

Raj Shankarappa

Copper Staining of Gels:

The method involves the soaking of freshly-electrophoresed SDS-PA gels in 0.3M CuCl2 for up to 15 min, washing the gels briefly in de-ionised water, and viewing them against a dark-field background: proteins come up as clear zones in a translucent blue background; the detection of proteins is intermediate in sensitivity between Coomassie and silver staining . Gels may be stored in water for up to several months at room temperature with no problem or fading: proteins are immobilised as a Cu-SDS- polypeptide complex in the gel , which remains clear; the colour and background opacity are due to a Cu-SDS-Tris complex. Gels may be destained completely by repeated washing in 0.1-0.25 M Tris/0.25 M EDTA pH 8.0, and then electroblotted, or eluted from the gel for other purposes.

NB: NATIVE PA GELS CAN ALSO BE STAINED : this is a less sensitive procedure, and bands are visible as faint blue opaque bands on a clear blue background.



Rybicki and von Wechmar, 1982; J. Virol. Methods 5:267-278;

Towbin et al. 1979, Proc. Natl. Acad. Sci. USA 76:4350.