4. When we created our imaginary example western blot, we expected to get equivalent amounts of the protein of interest in each lane, because we (hypothetically) loaded equivalent sample volumes of protein from the same vial of homogenate in each lane. During the loading process, something went awry, and we got very different densities on the western blot. But after carrying out the loading-control corrections above, we find that our Adjusted Densities for each of the 4 lanes are roughly equivalent (all are approximately = 1), indicating that there are equivalent ratios of the protein of interest contained within the sample of total protein in each lane, as we would expect.
The example above represents a special case. In reality, you will typically be skilled enough to pipette equivalent amounts of total protein into your gel lanes, presumably because you have used a process like the BCA Protein Assay to quantify the total amount of protein present in each of your homogenate samples. This can obviate the need for the loading-control bands and the associated controversy about whether the protein you’re using to assess equal loading is truly expressed consistently across all treatment levels.
Extending the example to multiple western blots
If you can apply equal amounts of total protein to each lane of a gel, then you only need to concern yourself with having an appropriate standard sample placed on each western blot you run for the project. For example, you might purchase an aliquot of purified Human Hsp70 and place 1μl in a lane on each gel you run when probing for Hsp70. Or you might save some money by creating your own standard sample by mixing aliquots of several of your homogenates to make a large quantity of mixed homogenate that you can use to load a standard volume onto a single lane of each gel for your entire project (running out of this mixture part way through your project would be a Bad Thing). In this case you may not know the true concentration of the protein of interest in your mixed homogenate, but you will at least be able to get an equivalent amount of total protein (including the protein of interest) on each gel.
The goal of the standard sample is to account for variations in antibody binding efficiency (for example if you’re re-using a mixture of antibody on multiple western blots to save money), for variations in blocking and washing efficiency, for variation in the decay rate of a chemiluminescent reagent used to visualize the antibodies, or for variation in the exposure time of your x-ray film. In any of these cases, the intensity of the signal you get out of your western blot can vary from blot to blot (Figure 21). You can use a standard sample on each gel to normalize every other band on that same blot, and then compare across multiple blots because every band in your dataset is normalized to the same standard (be it 10ng of Human Hsp70 or a mixture of several homogenates).
Finally, if you are using a loading-control protein band (lower row on these blots) on each gel, and a standard protein sample on multiple gels, you can correct both for differences in total protein loaded in a lane (using the loading-control band) and for differences in exposure between gels (using the standard sample).
For example, consider the two gels in Figure 22. The 1st (white) gel has minimal background noise, and there is some variation the amount of protein loaded in each lane, as revealed by the size and density of the loading control bands of the samples. The 1st lane on the left is our standard sample used on each gel. The 2nd (gray) gel has more background noise due to a different exposure time, poor washing, or any number of possible problems. Again, there are different protein quantities loaded in the 4 lanes, based on the size and density of the loading-control bands on the lower half. The 1st lane on the left is the same standard protein sample as that used on the 1st gel.
We would start by analyzing the two gels separately, using the protocol in the loading-control portion of this document.
1. Calculate the relative density of the loading-control bands (lower row on this gel). Use the loading-control band for the standard in lane 1 to do this calculation.
2. Calculate the relative density of the bands for the protein band of interest (upper row on this gel). Use the protein band of interest for the standard in lane 1 to do this calculation.
3. Adjust the relative densities calculated in Step 2 using the relative densities for the loading-control bands calculated in Step 1. Simply divide the relative density from Step 2 for each lane by the corresponding relative density from Step 1.