It is often important to be able to direct the antibody response to specific regions of the target protein. Reasons may include targeting functional regions of interest, discriminating small differences such as single amino acid substitutions or post-translational modifications. In addition, if a diverse set of epitopes are required to comprehensively cover the target protein then it is often better to fragment the protein up into domains and immunize separately to overcome immune dominance effects seen the whole protein.
We have explored this concept of targeted specificity with the model protein perlecan, a >4000aa heavily glycosylated protein composed of >30 domains that is subject to proteolysis at multiple sites. In order to develop sensitive sandwich immunoassays capable of detecting different proteolytic fragments we generated a panel of antibodies directed to different domains of perlecan.
The antibodies were evaluated in western blot and immunoprecipitation. Some of the antibodies performed well on western blot, some on immunoprecipitation , some on both, some on neither.
A select set of these antibodies were tested in a matrix of sandwich immunoassays performed on a microarray to identify the best performing pairs of antibodies. Different sets of pairs can be used to differentiate proteolytic fragments of a protein.
Frequently assays are required to discriminate between two closely related proteins, or detecting small post-translational modifications such as phosphorylation. Antibodies have the capability of detecting small differences between two highly similar proteins.
There are two general strategies used to generate antibodies capable of discriminating with such fine specificity; use a small antigen (hapten or a short peptide) to focus the immune response to the desired site, or use a large antigen and screen. Small antigens (eg. peptides) are in essence a simpler approach, but can have limited success if the desired immunoassay platform displays folded antigen since the antibodies from peptides usually perform poorly with folded protein.
Large antigens (eg. proteins) can be used to successfully generate antibodies to discriminate small differences and have the advantage of being more successful at generating antibodies that recognize folded antigen. However, there are two general difficulties:
1) ensuring the immune response focuses on the epitope,
2) isolating the desired antibody from amongst the background of antibodies to other epitopes. This problem can be minimized by selecting small stably folded regions of the protein centered on the difference (ie. domains).
Proteins then need to be generated to the undesired form to allow the antibodies to be screened to identify ones with the appropriate specificity.