Antigenic mapping and characterization of Albumin Binding Protein
(English)Manuscript (preprint) (Other academic)
The possibility to predict the location of antigenic determinants is a desirable feature in antibody production ventures and for vaccine development. However, antigenic propensity scales available today are poor, and so far it is not possible to predict the best antigen to trigger the immune system. Here, a unique set of 411 antisera towards a common part of allantigens within the Human Protein Atlas project has made it possible to perform massive epitope mapping. This effort generated a true map of the antigenic regions of this common N-terminal tag, and rendered it possible to further investigate what features that generate a good antigen. Investigations on variation in epitope occurrence are often an obstacle when mapping antigens, because of the ethics of using more animals than necessary for antibody production. As a consequence, not much has been done to verify epitopes found and the variance between different immunizations has not been thoroughly investigated. Herein it was shown that the most immunopotentating sites were only detected by the polyclonal antibodies in 70% of the immunizations, demonstrating the need of good antigen design. Detected epitopes also showed that aromatic amino acids, some positively charged aminoacids, and serine and glycine were over-represented in the antigenic hot spot regions. The detected antigenic regions were also shown to have fairly low correlation to several antigenic propensity scales.
IdentifiersURN: urn:nbn:se:kth:diva-25855OAI: oai:DiVA.org:kth-25855DiVA: diva2:360249
QC 201011022010-11-022010-11-022010-11-02Bibliographically approved