Repository logo
 

FROM ANTIBODY-DRUG CONJUGATES TO MASKED ANTIBODIES: BIOPHYSICAL INSIGHT FOR THE RATIONAL DESIGN OF FUTURE THERAPIES


Type

Thesis

Change log

Authors

Abstract

Monoclonal antibodies have become a major class of therapeutics over the past twenty-five years and, during this time, many successful antibody scaffolds have been based on wild-type humanized and human sequences of the immunoglobulin G isotype 1 (IgG1) subclass. In the past, several mutations have been engineered into these scaffolds, to optimise biological/physical properties or to allow site-specific conjugation. In the first project, the thermodynamic, thermal and kinetic stabilities of two sets of mutations were measured: one was a triple mutation in the CH2 domain (L234F/L235E/P331S), and the other had an additional substitution in the CH3 domain, S442C, to enable site-specific conjugation of a cytotoxic payload. Overall, results showed that the order of the domains in increasing stability were CH2, CH3 and Fab. The triple mutation was found to destabilize the CH2 domain, while the substitution in the CH3 domain did not affect stability significantly. Antibody-drug conjugates have been one of the most actively developed classes of drugs in the past fifteen years combining the strengths of large and small molecule therapeutics. Recently, strategies involving the insertion of a cysteine and maleimide linkers to achieve site-specific conjugation have been developed in order to attain highly controlled drug to antibody ratios. One such antibody scaffold, Fc-C239i, that formed an unexpected disulfide bridge during manufacture, was characterized. A combination of mass spectrometry and biophysical techniques were used to understand how the additional disulfide bridge forms, interconverts, and changes the stability and structural dynamics of the antibody. Alternative strategies to the use of maleimide linkers have recently been developed, e.g. DVP and tetraDVP linkers, by the Spring Group in the Department of Chemistry. The impact of the conjugation of DVP and tetraDVP-linkers on the stability and dynamics of the antibody trastuzumab was investigated. Results showed that the linkers destabilised the CH2 and constant domains of the Fab, to a degree which is very similar to those observed for other antibody-drug conjugate scaffolds. Although monoclonal antibodies have greatly improved cancer therapies, they can trigger vi side effects due to on-target off-tumour toxicity. Recently, strategies have emerged to mask the antigen-binding site of antibodies, such that they are only activated at the tumour site. Here, the underlying mechanisms that determine what makes an effective anti-idiotypic antibody fragment mask were investigated, using three masks with different properties. Four main parameters were established as playing key roles. The efficacy of inactivation relies, first, on the extent of binding site overlap with the antigen, and second, on a relatively high association rate constant for mask and antibody. The ease of activation relies on the antibody having a lower affinity and higher dissociation rate constant for the mask than for the antigen. Fourth, the closer the affinity of the mask and the antigen for the antibody, the more disruptive the activation step needs to be. Hydrogen-deuterium exchange mass spectrometry is a technique which probes molecular dynamics at high-resolution. To obtain high resolution, it is necessary to obtain a good peptide map. In order to understand the factors that limit the essential digest step and therefore the coverage in the peptide map, unfolding studies on an antibody were undertaken under the quench conditions used in HDX-MS experiments. Results provide confirmation that the rate of unfolding of an antibody domain limits resolution in these experiments.

Description

Date

2021-09-29

Advisors

Jackson, Sophie

Keywords

antibody-drug conjugates, masked antibodies, protein folding, protein stability, pro-antibodies, mass spectrometry

Qualification

Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge
Sponsorship
EPSRC (1783776)
EPSRC (Grant EP/L015889/1) and AstraZeneca