Choice of Human Frameworks:

Despite this being the most critical part of the antibody Design Cycle, there are no hard-and-fast rules for choosing the human acceptor frameworks into which to graft the donor CDRs. This is because the benefit of the various choices (in terms of immunogenicity in the patient) has not been clearly proved in the clinic. Therefore, there are only a set of approaches which need to be combined with the collective experience of previous humanisations.

Having decided on an approach to take in order to choose the human frameworks; which particular human antibody, consensus or germline sequence should be used? This is simple in the fixed framework approach since the choice is always NEW for the heavy chain and REI for the light (although since modifications to both frameworks are apparent from the literature (Campath-1), even there a choice needs to be made). The choice is usually made by performing a search for the most homologous sequence over the appropriate database. There are many database search programs available, the two most common being BLAST and FASTA. BLAST is often preferable since it allows searching over the Kabat database. In order to search the consensus and germline sequences, they need to be downloaded onto the local computer. Choice of the particular human frameworks for the light and heavy chain variable regions should be made by trying to match the length of the CDRs, the canonical residues and the interface packing residues (as described for the donor sequence) as well as trying to find the highest percentage identity between the donor and acceptor sequences. Unusual residues in the acceptor frameworks are not usually tolerated since they may lead to immunogenic epitopes in the resultant humanised antibody. Try to find human frameworks which are similar (in terms of percentage identity) to the source sequences and also require the least number of backmutations.

Home Page José Saldanha © 2000. Birkbeck College, London WC1E 7HX.