The study of the proteome poses a formidable challenge in selecting optimal expression systems capable of reliably producing assay-ready proteins. Drawing from personal insights in protein expression in ag and human biotech, this discussion explores effective approaches to protein production. Key considerations such as cost efficiency, protein localization, disulfide bonding, glycosylation, organism source, team dynamics, and downstream assay prerequisites will be examined. Presently, no single expression system can consistently deliver proteins of requisite quantity and quality for downstream assays. Biases stemming from familiarity with certain expression systems can inadvertently lead to the production and utilization of suboptimal protein forms. Ultimately, the discussion concludes with insights into the critical need for high-quality data to properly train and validate advanced machine learning and artificial intelligence models.

While many proteins are indeed best expressed in a specific expression system, researchers should remain vigilant and keep an open mind. Case studies will be presented that illustrate how the initial and 'obvious' choice of E. coli, insect, or mammalian (and Vibrio natriegens) expression system was inappropriate.

The drug discovery landscape is ever-evolving and constantly demands ways to produce difficult-to-express proteins in a fast and efficient manner. In our department, Biomolecular Research at Genentech, we have implemented several expression platforms to enable speedy and robust expression screening for structural and biochemical studies. This presentation will highlight some of the key challenges which we face when expressing a large amount of diverse intracellular and membrane proteins.

Transient mammalian cell transfection is a gold-standard method for production of various large-molecule modalities in the drug discovery setting. However, production of high-quality cell culture seed with =99% viability is typically a labor-intensive and hard-to-automate process, especially when 10-100L of culture is needed. Here, we describe our end-to-end automated cell culture platform to prepare, maintain, and deliver large volumes of high-quality cell culture.

Through gene-specific cell engineering, we have demonstrated the power of engineered CHO cells, (geCHO) enabling rapid production of bespoke glycoforms of therapeutic proteins. With this platform, we have produced and screened multiple therapeutic drug candidates (vaccines and enzymes) in vitro and in vivo to determine the optimal glycoform. Robust and efficient targeted engineering can be used to address specific challenges like contaminating HCP's, unwanted product activity, and more.

Since their establishment by Imogene Schneider in 1972, Drosophila melanogaster S2 cells have been a staple in academia for their ease of handling and versatile protein expression. Viral methods or induction of expression can reduce cell viability, posing challenges for manufacturing due to process heterogeneity up- or downstream. High-viability cell production enhances product homogeneity and batch consistency, essential for API manufacturing. ExpreS2ion Biotechnologies has advanced the use of S2 cells for APIs, achieving Phase III clinical validation with development of a COVID-19 vaccine. We continue to develop the ExpreS² system, including enhancing glycosylation pathways to modulate product immunogenicity.

This study explores the potential of the fungus Thermothelomyces heterothallica C1 as an alternative cost-efficient platform for monoclonal antibody (mAb) production. We have engineered the glycosylation pathways of C1 to produce human-like N-glycans with good productivity and product quality. Production levels over 9 g/l of secreted mAb with desired glycan profile have been obtained.

The expansive genetic toolkit for Drosophila melanogaster allows for stable expression of bioactive proteins. Illustrated through multiple case studies, we will discuss genetic strategies on optimizing protein expression, including tissue-specific and inducible expression. This novel approach has the potential to overcome recombinant protein expression difficulties associated with conventional systems.

Saccharomyces cerevisiae is widely used for recombinant protein production but often yields lower protein levels compared to other systems. To identify key bottlenecks, we developed two systems-wide approaches by screening a genomic library of over 4,000 mutant strains and ~1,000 isolates from diverse environments to find strains producing higher levels of a heterologous laccase. Gene ontology analysis revealed enrichment in processes like vesicle trafficking, vacuolar degradation, and metabolism. We validated that deleting several of these genes enhanced recombinant protein production in our lab strain, suggesting new strategies for strain bioengineering.