Application of antibodies as highly specific binding targets has sofar been limited to the pharmaceutical/medical area, primarily as targeting agents. Whilst there exist ample possibilities outside this area, such as e.g. targeted antibacterial components for highly effective preservative functions, reducing environmentally unfriendly chemicals, applications outside the traditional field are prohibited by the low availability and stability of these proteins. This project will apply the techniques of protein engineering and protein design to antibody fragments in order to make them more suitable for large scale industrial production and applications.
In particular protein flexibility and domain interactions of antibody fragments will be studied in order to obtain general rules for increasing protein stability and `robustness' in industrial processes. At the same time, the influence of protein internal dynamics and flexibility on the binding of antigen will be studied. Furthermore, the design of stable mini-variants of antibodies, consisting only CDR(s) will be examined by reversed 'PEPSCAN'. X-ray structures of two antibody fragments will serve as the basis for designing variants of the proteins. The X-ray structure of the antibody McPC603 has already been determined, making this the model system to test general hypotheses. The X-ray structure of the antibody M02/05/01, directed against the commercial musk odourant `traseolide', will be defined. On the basis of these structures and extensive molecular modelling experiments a set of variant molecules will be made at the hydrophobic core of the protein aiming to optimize packing and tighten interactions in this region. We expect more stable variants to emerge from this approach. On the surface of the protein near the antigen binding site, we will introduce metal binding sites, in order to influence the mobility of surface loops. Effects of these, especially also on binding behaviour, will be carefully studied by spectroscopy and physico-chemical characterization techniques. With the help of molecular modelling, attempts will be made to extract general design rules from the results. The two proteins serve to cross-test these rules.
Variants of the anti-phosphocholine and anti-traseolide antibody have been made that are more stable and show much higher production levels. The X-ray structure of the Fab fragment of the anti-traseolide antibody has been determined, crystallization of stable variants of the anti-phosphocholine antibody only yielded microcrystals unsuitable for X-ray experiments. Based on a homology model of the anti-traseolide antibody a 17 residue peptide was designed that specifically binds traseolide with an equilibrium dissociation constant only a factor 30 higher than the original antibody. New improved tag sequences with differential and selective recognition properties for the anti-FLAG antibodies M1, M2 and M5 were identified. A naturally occuring anti-traseolide antibody lacking the conserved disulphide bridge in the VH domain was cloned as a scFv frgament and could be expressed as a functional protein in the periplasm of E. coli. Which is in contrast with the general believe that the disulphide bridges in both variable domains are essential for proper function of antibody fragments.
MAJOR SCIENTIFIC BREAKTHROUGHS:
The high resolution structure of the Fab fragment of the anti-traseolide antibody in complex with and without traseolide has been determined. A set of stabilizing/production level enhancing mutations for phosphocholine and traseolide binding has been identified. A peptide has been designed starting from a homology model that binds specifically traseolide. Improved FLAG sequences for the anti-FLAG antibodies M1, M2 and M5 have been found, which can be used for antibody fragment (protein) purification and characterization. An anti- traseolide antibody fragment lacking the highly conserved disulphide bridge is expressed as a functional protein in E. coli.
Funding SchemeCSC - Cost-sharing contracts
8200 AB Lelystad