Historically purification process development leveraged a few key macroscopic parameters including class of biologic molecule (therapeutic protein – MAb or other, natural product, vaccine), pI, and molecular weight to guide purification development options. This limited data set coupled with empirical approaches, including high throughput screening and analytics, allowed for development of a purification process. A directed approach based on a more detailed biophysical characterization is required to ensure successful process development to support accelerated development with a more diverse range of molecules. To achieve this goal, our current research focus is on understanding the biophysical properties of each specific biologic molecule in silico at an atomistic level. The approach, combining docking with molecular dynamics, has initially been applied and qualified vs. experimental data for chromatography (IEX, RP and Mixed Mode) for small proteins (ex. insulin and associated variants). This research in this paper demonstrates the applicability of this approach to complex systems (including Mabs) with a range of ligands to explore the impact of specific clusters (electrostatic, hydrophobic or mixed mode) on purification modality choice. The potential impact of avidity on binding affinity and selectivity are also examined in detail. Additional insights into chromatographic performance for a specific ligand were obtained via a comparison of interaction energies for highest affinity sites on the MAb and Dynamic Binding Capacities for a specific ligand. The potential impact of indirect solvent mediated events on protein/ligand binding was assessed via a direct comparison of implicit and explicit solvents and counter-ions in the model. Leveraging collaboration with Discovery, this approach allows for rapid assessment of feasible purification modalities to resolve specific molecular liabilities as well as potential fit to platform assessment. This collaborative in silico assessment allows for rapid development of product candidates and can be applied to biologics and can be readily applied to both chromatographic and non-chromatographic purification of biologics. Thus, successful implementation of this computational model at the onset will significantly enhance the development of biologics from discovery, through purification to formulation.

Rationale purification development: Leveraging molecular biophysics to target purification modalities

249th National Meeting (2015)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.