Llama heavy chain antibodies, to mutate or not to mutate (THESIS VERSION)

Abstract

A high production yield and good stability are required for the large scale application of llama heavy chain antibody fragments (VHHs) against targets such as a microbicide against HIV. Many aspects of the production process are amenable to optimization by engineering. Codon usage, for example, can be improved, but it is also possible to remove proteolytic cleavage sites, chaperone binding sites, or to improve the packing of the protein by increasing the numbers of hydrophobic contacts or hydrogen bonds. However, not every amino acid can be mutated freely. Obviously, the cysteine bridge should stay intact at all cost. It also seems unwise to mutate residues involved in scaffolding the CDRs. We have made a large series of in silico studies, and tabulated and analyzed all known mutations in VHHs. No mutation information was available for eight residues that were predicted to be important for folding and production yield using the rules for in vitro folding of Vendruscolo. When we mutated these eight residues we observed thermostability differences ranging from -1 to -14 degrees in agreement with what one would expect from the selected modifications. The production yield of most of the mutants decreased as well. All known and predicted mutation effects were combined in a generic 'VHH-mutability' table. This table could be used to flawlessly predict the best producing variants when a series of HIV neutralizing VHH variants was obtained.