Antigen Prediction

Antigen Prediction

Antibodies are used to detect antigens in a natural environment and in order to do so, they must have access to the part of the antigen for which they have specificity. Proteins, being three dimensional structures, will have regions completely inaccessible to antibody, for example within the centre of a globular protein structure. An antibody is a large molecule and may be unable to penetrate the matrix.

To raise an antibody that can bind to the native protein, the peptide must be selected from an accessible region of the protein if the resulting antibody is to be of use. The most accessible areas will be those parts of the molecule that are exposed or on the outside of the structure. As these regions are in contact with an aqueous environment they are usually hydrophilic.

The peptide should also adopt a conformation that mimics its shape when contained within the protein. Finally, the peptide must be immunogenic. So, how do you go about selecting antigenic peptides?

Antigenic prediction

Hydrophilic regions of the protein are usually exposed in aqueous solvents and are thus available for antibody recognition. Computer programs are available within most protein database systems to produce a hydrophilicity plot for the protein. This is done by assigning a “hydrophilic index” to each amino acid in a protein and then plotting out the profile in which the regions of hydrophilicity can be seen. From these regions we can select a number of potential peptide antigens for synthesis, but the final choices may be influenced by other considerations. For example, antibodies may be required to recognise post-translationally modified regions (eg phosphorylation or glycosylation sites) of the target protein and the chosen peptides would need to span these regions. On the other hand, sites which can undergo post-translational modification will need to be avoided if antibodies to the unmodified protein are required. Transmembrane regions should be avoided because these may not be accessible for immunodetection by the antibody.

Choice of Peptide Sequences

Peptides of 15-20 amino acids are normally selected. A single antigenic determinant or epitope is between 5 and 8 amino acids and therefore a peptide of 15-20 amino acids will contain one or more epitopes. Longer peptides may have a greater conformational similarity to the native protein and be more likely to induce antibodies that recognise the natural protein. Clearly, any peptide selected must be capable of successful synthesis. It must also be readily soluble in an aqueous buffer for conjugation and for use in biological assays.

If a hydrophilic region has been selected then peptide solubility should not be a problem. However, even these regions may contain hydrophobic residues (e.g. trytophan, valine, leucine, isoleucine and phenylalanine) and if there is a choice, it is better to select a peptide with as few of these residues as possible. Glutamine may also cause insolubility as it can form hydrogen bonds between peptide chains, so multiple glutamines are to be avoided.

A cysteine in the selected sequence is useful for conjugation of the peptide to the carrier protein. The cysteine should be at the N or C terminus for optimal antibody response to the sequence. However, two or more cysteines should be avoided because disulphide bonds may form within and between peptide chains leading to insolubility and structural alteration. When the chosen peptide lacks cysteine, a cysteine residue can be added to the sequence at the N- or C-terminus to facilitate conjugation to the carrier.

Proline can adopt a cis-amide bond structure (normally in peptides amide bonds are trans) and consequently it can give the peptide a shape that may mimic more closely the shape of the peptide in the protein. Normally, peptide chains tend to be random in structure and the introduction of a proline can induce structural motifs thereby enhancing its potential as an immunogen.

The selection process will finally reduce the candidates to 3 or 4 peptides. If possible at least 2 and preferably 3 peptides should be selected and synthesised. This greatly increases the chance of at least one peptide being successful in providing an antibody to the target protein.

The purity of the peptide used for antibody production will greatly influence the quality of the antibody response. The purer the peptide, the greater will be the specific antibody response to the target.

CRB provide a comprehensive antigen selection service and we would be pleased to advise on the appropriate peptides to synthesise.