Therapeutic Antibody Engineering To Improve Viscosity and Phase Separation Guided by Crystal Structure

Abstract

Antibodies at high concentrations often reveal unanticipated biophysical properties suboptimal for therapeutic development. The purpose of this work was to explore the use of point mutations based on crystal structure information to improve antibody physical properties such as viscosity and phase separation (LLPS) at high concentrations. An IgG4 monoclonal antibody (Mab4) that exhibited high viscosity and phase separation at high concentration was used as a model system. Guided by the crystal structure, four CDR point mutants were made to evaluate the role of hydrophobic and charge interactions on solution behavior. Surprisingly and unpredictably, two of the charge mutants, R33G and N35E, showed a reduction in viscosity and a lower propensity to form LLPS at high concentration compared to the wild-type (WT), while a third charge mutant S28K showed an increased propensity to form LLPS compared to the WT. A fourth mutant, F102H, had reduced hydrophobicity, but unchanged viscosity and phase separation behavior. We further evaluated the correlation of various biophysical measurements including second virial coefficient (A2), interaction parameter (kD), weight-average molecular weight (WAMW), and hydrodynamic diameters (DH), at relatively low protein concentration (4 to 15 mg/mL) to physical properties, such as viscosity and liquid-liquid phase separation (LLPS), at high concentration. Surprisingly, kD measured using dynamic light scattering (DLS) at low antibody concentration correlated better with viscosity and phase separation than did A2 for Mab4. Our results suggest that the high viscosity and phase separation observed at high concentration for Mab4 are mainly driven by charge and not hydrophobicity.

Beyond CDR-grafting: Structure-guided humanization of framework and CDR regions of an anti-myostatin antibody

“…In silico optimization can be used for many other properties of antibodies required for successful development. Properties such as viscosity, aggregation propensity, chemical instability, phase separation and reduced clearance rates benefit from a co-crystal structure for direct prediction, [73][74][75][76] as well as to prevent the introduction of destabilizing and affinity reducing mutations. Additionally, as described previously, determining the epitope of interaction can be useful in understanding the biological activity and function of the antibody.…”

Mitigation of reversible self-association and viscosity in a human IgG1 monoclonal antibody by rational, structure-guided Fv engineering

“…9,15,[18][19][20] In addition to the multitude of distinct domain surfaces that can compose the binding interfaces, the nature of the driving forces behind antibody selfassociation can also be diverse, with both electrostatic and hydrophobic interactions playing key roles. 6,7,18,21,23 Potential strategies to mitigate self-association and other undesired high concentration solution properties may include protein engineering approaches, 20,[24][25][26] as well as formulation development and optimization efforts. [5][6][7]9,18,19,21,23,27 While the former can be limited due to the degree of knowledge of sites of interactions, the latter can be time and resource intensive and may not always be fully successful to a desired degree for all antibodies.…”

Computational investigation of protein surface polarity and pH-dependence: application to antibodies shows that antibody-antigen interfaces tend to be relatively polar

“…The current work provides an additional tool for groups looking to identify regions of a protein for engineering improved solution properties. Both hydrophobic patches (Nichols et al, 2015), and charged patches (Yadav et al, 2012;Chow et al, 2016;Perchiacca et al, 2012) have been mutated to alter solution behavior. The ability to rapidly visualise surface patches will further inform and accelerate such work.…”

“…Much research has focused on the role of anisotropic surface patches of charge and hydrophobicity in causing reversible and irreversible protein association (Yadav et al, 2012;Perchiacca et al, 2012;Chow et al, 2016;Li et al, 2016;Roberts et al, 2014b;Austerberry et al, 2017). This experimental work has led to efforts in predicting protein surface patches in silico.…”