(27bi) Streamlined Yeast Cell Reactors with Residence Time Control to Engineer and Profile Protein-Modifying Enzymes

The ability to engineer and profile the substrate specificity of protein-modifying enzymes and to develop targeted therapeutics against them is central to biotechnology, chemical biology, and biomedicine. In this vein, high-throughput platforms that empower these experiments are key to advancing the field. Developed ten years ago, the yeast endoplasmic reticulum (ER) sequestration and screening (YESS) system remains a powerful method to quantify catalytic turnovers of protein-modifying enzymes and to select for desired activities by yeast surface display¹. Originally developed to engineer and profile protease activity, in the YESS system, a protease and its target substrate are co-expressed in the yeast ER. The substrate cassette is coupled with the yeast mating factor agglutinin A protein subunit, Aga2p, which allows for display on the yeast surface^2,3. The system takes advantage of two epitope tags FLAG and HA, which flank a substrate sequence. As the AGA2 substrate cassette travels through the ER, if the protease successfully cleaves the substrate, then the HA tag is lost. Subsequently, when yeast cells harboring YESS plasmids are stained with fluorescently labeled anti-FLAG and anti-HA antibodies, the ratio of FLAG to HA fluorescence on the yeast surface translates into protease activity. In principle, YESS could be used for any protein-modifying enzyme whose activity can be detected on the yeast surface, orthogonally from the endogenous enzyme activities. Expectedly, the YESS system has been expanded to study kinases and histone acetyltransferases.

Here, we present a toolkit that fully unleashes the potential of this system. First, we address the modularity of the YESS system, overcoming the design-build-test bottleneck. The YESS system contains at least two transcriptional cassettes: an enzyme and a substrate. The enzyme cassette is comprised of a promoter, an ER targeting signal, the enzyme, and an ER retention signal (ERS). Similarly, the substrate transcriptional unit contains a promoter, the substrate cassette (two epitope tags plus a substrate sequence), and an ERS. Assuming the enzyme is the only constant, this leaves at least 5 system parts to optimize when investigating enzyme activity. These include the right combination of ER retention signals on the protease and substrate, the choice of promoters to drive enzyme transcription, and the choice of substrate. Finally, one must decide whether the system should be fully or partially integrated in the yeast chromosome. Taking all of this into consideration, we built a fully modular YESS system where any combination of parts can be assembled via a BsaI Golden Gate into centromeric or integrative receiver plasmids. To showcase the utility of this toolkit, we mapped the activity landscape of proteases with respect to ERS combinations and enzyme transcription levels. Results from this experiment serve as a proxy for ER receptor engagement and saturation. This collection of parts will be made available on Addgene and should help the research community interested in engineering protein-modifying enzymes.

High throughput display methods that engineer the specificity of binders typically perform rounds of sorting in the presence of a significantly high concentration of a counterselection molecule. While YESS enables one to introduce both a counterselection and a selection substrate, which is paramount to fine-tune protease specificity. The current version of YESS does not allow one to vary the ratio of counterselection to selection substrate. To remedy this issue, we leverage the fact that ERS strength correlates with ER residence time and the fact that relative ER residence time can be measured on the yeast surface. We showed that integrating two substrate cassettes, one with a strong WEHDEL ERS, and another with no ERS, results in the expected high display of the no ERS substrate relative to the one with the strong ERS. Therefore, a substrate cassette with a strong ERS has a longer ER residence time than one with a weaker ERS or without an ERS. Effectively, an enzyme in the ER will “see” more of a substrate with a strong ERS, which can serve as a method to expose the enzyme to a high concentration of a counterselection substrate relative to a selection substrate. We tested this approach with an engineered TEV protease with specificity towards glutamate (TEV-E), rather than glutamine (TEV protease wild-type or TEV-Q), at the P1 position. TEV-E was obtained in the traditional YESS system where both the counterselection and selection substrates were on the same AGA2 fusion polypeptide. By using the new differential substrate display approach, we showed that TEV-E is not an entirely orthogonal protease variant, which precludes its full implementation in building synthetic protein circuits.

Most importantly, this final product, a Golden Gate-able YESS system, can easily be adapted to test completely new proteases by swapping out at most three parts in the system: the promoter, the substrate, and the protease. This highly modular tool will enable us to highjack a variety of proteases to leverage their immense physiological impact to improve human health.

References

1. Yi, L.; Gebhard, M. C.; Li, Q.; Taft, J. M.; Georgiou, G.; Iverson, B. L. Engineering of TEV protease variants by yeast ER sequestration screening (YESS) of combinatorial libraries. Proc Natl Acad Sci USA 2013, 110(18), 7229-7734. DOI: 10.1073/pnas.1215994110

2. Zahradník, J.; Dey, D.; Marciano, S.; Kolářová, L.; Charendoff, C. I.; Subtil, A.; Schreiber, G. A Protein-Engineered, Enhanced Yeast Display Platform for Rapid Evolution of Challenging Targets. ACS Synthetic Biology 2021, 10 (12), 3445-3460, DOI: 10.1021/acssynbio.1c00395

3. Denard, C. A.; Paresi, C.; Yaghi, R.; McGinnis, N.; Bennett, Z.; Yi, L.; Georgiou, G.; Iverson, B. L. YESS 2.0, a Tunable Platform for Enzyme Evolution, Yields Highly Active TEV Protease Variants. ACS Synthetic Biology 2021, 10 (1), 63-71, DOI: 10.1021/acssynbio.0c00452