Strain improvement for increasing protein production in S. cerevisiae

Strain engineering for protein production

When it comes to strain improvement for achieving the best protein yields, simply increasing the copy number or strength of the target protein expression cassette often does not cut it. Strain engineering requires a comprehensive approach that combines improving the target protein expression along with optimization of the genetic background of the host. The traditional methods of strain improvement – such as classical mutagenesis – can get you to the end goal but run up long development times and carry the risk of accumulating mutational load with potentially unpredictable downstream effects. 

Combining genome-wide and targeted approaches to strain improvement

The Onyx® platform provides a workaround for these problems: in just one experiment, you can generate thousands of genome-wide, trackable edits that allow you to see how the engineered genotype affects the strain performance. It lets you simultaneously leverage existing knowledge of targets shown to improve protein yields – for example, knockout of genes involved in protein degradation – and explore additional hypotheses by sampling the genome-wide sequence space. 

This is exactly what we did for improving production of cellobiohydrolase I (CBH1), an enzyme involved in the degradation of lignocellulose that is being actively explored for consolidated bioprocessing applications. We expressed the codon-optimized fungal CBH1 gene is S. cerevisiae with the corresponding native signal sequence under the control of a yeast ENO2 promoter. Then, we designed several concept libraries spanning diverse targets and edit types, for a total of 14,859 edits across the genome. 

Protein Production Figure 2 C
Activity improvement in edited strains across different libraries.
Screening CBH1 libraries for improved activity

After screening several thousands of isolates from different libraries, we identified 74 unique hits that improved production of CBH1 in our strain. These included previously known and novel targets across a wide range of functions: protein degradation and secretion, stress response, transcription and translation, nuclear transport, and uncharacterized genes. These hits came from both rational and exploratory library types, including transcription factor binding site and terminator engineering, genome-wide knockouts and short deletions, targeted knockouts, and alternate codon libraries. 

To learn more about the insights from this study and how these can be leveraged for improving your protein production strains, read the app note or listen to the webinar. 

Resources support logo

Resources & Support

Want to learn more about the Onxy platform and what we’ve been up to? Check out our latest videos, webinars, and presentations and get answers to our FAQs on the Resource & Support page.