Characterization of the folding and assembly of single-chain antibodies

Date
2011
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
The focus of this thesis is to study the refolding of single-chain antibodies (scFvs), 4M5.3 and 4-4-20, expressed as inactive inclusion bodies (IBs) to their active monomer form. Previous work on the chemical refolding for 4M5.3 revealed two monomer forms, where one was less active (A1) than the other (A2) as determined by fluorescence quenching assays. Numerous biophysical assays were employed to determine structural differences, which may explain the difference in binding characteristics. The biophysical assays and examples in literature pointed to a cis/trans isomer difference at proline residue 100 for A1 (trans form) and A2 (cis form). Numerous mutational studies were conducted to stabilize the cis form of Pro100 in 4M5.3 but none of the mutants generated only one form of the molecule, suggesting that there maybe a different proline residue responsible for the difference in the structure (hydrodynamic radius and surface hydrohpbicity) of A1 and A2 or some other structural reason for the difference, such as differences in salt-bridging, hydrogen bonding, hydrophobic interactions or a combination of all these factors at the interface between the light and heavy domains of the scFv. A different method for refolding 4M5.3 using hydrostatic pressure from inclusion bodies, which does not use any denaturants, also, showed A1 and A2 after purification by size-exclusion chromatography showing that the formation of the less active form is independent of the refolding method. The use of hydrostatic pressure to refold scFvs from IBs was extended to another scFv, 4-4-20, and to the addition of their ligand, fluorescein, during refolding in order to study its effect on the refolding process. The inclusion of ligand increased yields of scFv by two-fold for 4-4-20 and by four-fold for 4M5.3, which is the higher affinity mutant for fluorescein. Two methods were investigated to obtain the free form of the scFvs by removing the bound ligand. A controlled-dilution and filtration method which uses a low concentration of denaturant to dislodge the ligand and then filtration with renaturation buffer to flush the denaturant away was found to yield the highest amount of free active monomer for 4-4-20 and 4M5.3. However, the scFvs produced by this method were found to be of lower quality, where they had a slower on-rate and a faster off-rate for their ligand, as compared to chemical refolding. The final chapter of this thesis investigated ways of measuring the macromolecular folding rate of each scFv and, also, the folding rates of each domain and interface for each scFv. The determination of the rate of refolding for 4M5.3 and 4-4-20 by dilution and measuring the rate of fluorescence quenching was successful. This method would be useful for measuring the rate of refolding for very small amounts of protein or if the tryptophan (Trp) residues in the molecule were removed, since there was a way to track the rate of active protein appearance from unfolded protein. Trp to phenylalanine (Phe) mutations were made to 4M5.3 and 4-4-20 with the strategy of creating mutants that would have a fluorescent Trp residue in the domain or interface of interest as a tracker for the rate of folding for that domain. Trp mutants for each scFv were successfully created as verified by DNA sequencing and their activity verified by fluorescence quenching assay. However, we were unsuccessful at measuring a rate of refolding for the domains and interfaces of 4M5.3 and 4-4-20 using a rapid dilution method and tracking Trp fluorescence. The rapid dilution method was unsuccessful at refolding adequate amounts of refolded protein to detect enough signal from Trp fluorescence.
Description
Keywords
Citation