Authors

Discussion

Discovery Sciences at Astrazeneca is responsible for identifying molecules active against novel targets in a range of therapeutic areas. For most drug discovery projects in this early phase it is integral to have high quality recombinant protein to drive screening and lead optimisation activities.

Here we describe our platform and processes for recombinant protein production at AZ. This comprises a number of stages:

1. Construct design and molecular biology- currently driven by literature, bioinformatics and personal experience but our ambition is to move towards machine learning and AI driven construct design. Molecular biology is currently outsourced to an AZ sister team in Sweden.

2. Protein expression- our focus is predominantly on disease relevant human proteins. In order to support their expression we use a host of systems (E.coli, SF21 insect cells and HEK293/CHO mammalian cell lines). For all constructs we carry out expression screening at a small scale following a standard protocol.

3. Purification- following selection of the most promising & relevant constructs, they are grown at a larger scale for purification. We use ÄKTA systems to purify proteins in a semi-automated manner using established chromatography techniques such as Affinty and Size Exclusion Chromatography.

4. Quality control- We employ a host of QC techniques depending on the end use of the protein. At a minimum this is SDS-PAGE, Mass Spec, and Size Exclusion Chromatography.

We are constantly looking for ways to increase throughput and streamline our methods to improve efficiency as we support an increasingly complex project portfolio and are stretched to reduce timelines.