Direct reprobing with anti-beta-actin antibody as an internal control for western blotting analysis.

Western blotting analysis is the most common method for identification and quantification of a specific protein in complex protein mixtures (2–4). To compare different levels of an interesting protein derived from different sources, such as from different tissues or from cell cultures exposed to a variety of treatments, an internal control is necessary to ensure even sample loading and gel transfer. Internal controls normally are housekeeping gene products because their expression levels are the same in most tissues and are presumably not affected by most treatments. Among these, β-actin is one of the most common proteins used for this purpose. There are at least two methods for performing the internal control analysis. The membrane that is used to detect the target protein is stripped and reprobed by another antibody against the internal control protein. This procedure alone could take two days. Also, the stripping procedure tends to elute proteins from blots that result in progressively decreasing signal intensities, and the binding affinities of proteins to membrane and/or antibodies to target proteins can be different, which results in poor reliability. Occasionally, duplicate membranes, one for detecting the target molecule and one for the internal control analysis, are generated under identical conditions. However, this strategy is more costly and still does not eliminate possible pipetting and transferring errors. We found a simple method to perform the internal control analysis in which the blot is directly reprobed by the internal control antibody after detection of the primary target. The major idea for this method is based on the theory that the binding sites of the target protein and its primary antibody are saturated, and horseradish peroxidase (HRP) conjugated on secondary antibody has been inactivated totally by hydrogen peroxide (H2O2) in substrate during the first detection (1). This method essentially avoids protein loss from the harsh stripping. Therefore, the detected internal control signal truly reflects protein loading. In our experiment, cyclooxygenase 1 (Cox 1) was the target protein, and βactin was used as an internal control. Thirty micrograms of total lysate of rat liver tissue from different treatments were separated on a 7.5% SDS-PAGE gel, then transferred to a piece of 0.2 μm nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA, USA). This membrane was blocked with blocking solution (5% non-fat milk, 0.05% Tween® 20 in PBS) overnight at 4°C with gentle shaking. The following morning, this membrane was rinsed once with PBST (0.05% Tween 20 in phosphate-buffered saline), then incubated with goat anti-rat Cox 1 M-20 polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) with a dilution of 1:500 in blocking solution. This incubation was performed at room temperature for one hour with gentle rolling in a hybridization tube. After the primary antibody incubation, the membrane was rinsed twice with PBST, followed by two 5 min washes and two 10 min washes to eliminate nonspecific binding. The membrane was then incubated at room temperature for one hour with 1:2000 dilution of donkey anti-goat IgG secondary antibody conjugated with HRP (Santa Cruz Biotechnology). The wash step was as described before, except an additional 10 min wash was added. After incubating with SuperSignal West Pico Chemiluminescent Substrate (Pierce Chemical, Rockford, IL, USA) at room temperature for 5 min, the membrane was wrapped with a piece of plastic and exposed to an X-ray film (Eastman Kodak, Rochester, NY, USA) for imaging. The Cox 1 blot membrane was kept at room temperature within the plastic wrap overnight. It was removed from the plastic wrap the next morning and washed once with PBST for 10 min at room temperature, then reprobed with a mouse monoclonal anti-β-actin antiBenchmarks