Antigen Design
Introduction:
To achieve optimal results in a custom antibody production project, it is essential to carefully design an antigen that will not only elicit a strong immune response during immunization, but that will also allow the antibody to bind to the protein in its intended application. Although the subject of antigen design is complicated and worthy of much greater detail than we can provide here, we have found several key principles that are essential for success in any application. As always, please contact our peptide specialists if you have any questions.
Intended Application:
Before beginning a new project, it is essential to understand some basic features of the protein of interest. In particular, knowing the protein's structure can greatly aid in choosing epitopes that are readily accessible to the antibodies. However, even without such detailed knowledge, the intended research application can affect peptide design strategy. For example, if the research focuses on various protein domains such as the C and N-terminals, or a particular state of the protein, such as phosphorylation, then peptides, and consequently antibodies, to the desired sequence can be used without much difficulty. However, protein conformation may interfere with antibody access to the target epitopes. The problem in this case is that a particular sequence may become inaccessible if it is hidden within the interior of the folded protein.
Epitope Selection Strategy:
In general, ideal antigenic epitopes are hydrophilic, surface orientated and flexible1. This is recommended because in most natural environments, hydrophilic regions tend to reside on the surface of proteins, while hydrophobic regions tend to be found hidden in the interior of the protein. Similarly, antibodies can only bind to epitopes found on the surface of proteins and tend to bind with higher affinity when those epitopes are flexible enough to move into accessible positions.
Continuous versus Discontinuous Epitopes:
Antibodies target continuous epitopes - epitopes that represent a continuous sequence of amino acids. Antibodies will bind to these regions with high affinity provided that the sequence is not located within the protein's interior. In some cases, antibodies against discontinuous epitopes (epitopes that represent a fold in a peptide sequence or even a joining of two separate peptide chains) can be generated, although this requires that the peptide used for immunization has a secondary structure similar to the epitope and that the sequence length falls within the guidelines listed below.
General Recommendations:
To counter the risk of selecting an epitope that is buried within the protein, we normally recommend that antibodies be generated against the C-terminus or the N-terminus, as these are often exposed on intact proteins. However, it is important to note that the C-terminus sequences on membrane proteins are often too hydrophobic to act as an antigen.
Sequence Length:
In general, we advise our clients to design peptide sequences between 8 and 20 amino acids in length. Any shorter, and there is a risk that the peptide will be so specific that antibodies would not recognize the native protein with sufficient affinity. Likewise, sequences over 20 amino acids in length risk losing specificity as well as introducing the potential for secondary reactions. Also, longer peptides are generally more difficult to synthesize at higher purities.
Carrier Protein Coupling Considerations:
Although coupling strategies can vary widely depending upon the sequence, it is important to remember that N-terminal sequences should be coupled at the C-terminal amino acid, just as C-terminal sequences should be coupled at the N-terminal amino acid. For sequences located within the protein's sequence, coupling can be done at either end. In general, the peptide should be linked to the carrier protein via the carboxy or amino-terminal residue. If no internal cysteines are present, then a cysteine should be added to the sequence. For more information, please contact our peptide specialists...
Software Solutions:
Several vendors offer software solutions that incorporate one or more of several accepted algorithms for predicting peptide antigenicity. Some of the more well known are:
http://www.accelrys.com/products/macvector/
http://www.dnastar.com/
References
1) Van Regenmortel, M.H.V., 1986, Trends in Biochemistry, 11:36-39