An in vivo assay for the high-throughput analysis and directed evolution of biopharmaceuticals

Directed evolution is a robust and powerful tool for engineering new and/or improved functions in biomolecules for therapeutic and industrial applications, as well as to uncover fundamental insights into protein behaviour. It works by exploiting the principles of natural evolution and accelerating it through multiple rounds of gene diversification and selection. In order to evolve the desired property, an appropriate assay for the property of interest must be chosen. However, improving proteins often proves challenging as most mutations are destabilising.

An in vivo TriPartite β-Lactamase Assay (TPBLA) has been shown to rapidly and easily identify misfolded, unstable or aggregation-prone sequences without the need for protein purification. Furthermore, TPBLA has successfully been utilised as a directed evolution screen to evolve thermodynamic stability and aggregation-resistance in a protein of interest. However, the methodology was limited in throughput, and due to the use of first-generation sequencing techniques identification of improved variants was laborious and high-cost. Chapter 3 in this thesis develops a new methodology for combining TPBLA with next-generation sequencing to enable high-throughput identification of hotspot regions and improved variants. This new approach has the potential to assess hundreds to thousands of variants in a single experiment and give a more comprehensive and extensive overview of a proteins' fitness landscape. In Chapter 4, this new high-throughput methodology is applied to biopharmaceutically-relevant targets to improve their aggregation behaviour. Furthermore, the ability of TPBLA to screen and rank a panel of clinically-relevant antibody therapeutics based on their developability is assessed. This demonstrated the potential of TPBLA for identifying poorly developable candidates, as well as potential late-stage clinical failures early in development prior to protein purification.

A common challenge for directed evolution studies is that often there is a trade-off between particular properties, such as stability and function, and by selecting for one you can negatively impact the other. Therefore, in the absence of a selection for function, evolving biopharmaceutical test proteins using TPBLA to improve their aggregation resistance could result in evolved variants that no longer bind to their target. Therefore, the work in Chapter 5 develops a novel assay, Solubility ‘n’ Affinity Coselection (SnAC), which introduces a selection for binding into TPBLA evolution experiments to enable evolution of biologics for both stability and function.

Overall, the work presented in this thesis details a novel and powerful approach for the analysis and directed evolution of stability and binding in biopharmaceutically-relevant proteins.