Conclusion and Discussion

           Ultimately, we successfully established nonconventional yeast production chassis for two kind of difficult-to-detect substances and attempted to build substrate-specific biosensors within them. By transplanting prokaryotic systems into eukaryotic cells and enabling their normal functioning, we expanded the application scope of biosensors. Furthermore, in the nonconventional yeast production platform, we constructed and tested the continuous directed evolution system, nyEvolvR, further transferring this system from Escherichia coli to yeast, especially nonconventional yeast strains with unique properties, thereby broadening its potential applications. In summary, we have developed an integrated eukaryotic yeast chassis that combines production, detection, screening, and directed evolution, providing an effective paradigm for the directed evolution of synthetic enzymes for various difficult-to-detect substances.

           At present, our sensor is still not sensitive enough to detect unless the presence of higher concentrations of borneol, that is, its detection limit is too high. The production of monoterpenes in microbial cell factories is still at the microgram level, and its production is difficult to be reflected by our detectors. We believe that the reason may be the leakage of operon, the insertion of the manipulation sequence may destroy the promoter or the insertion site is not the best. In order to solve these problems, increasing the copy number of the operator sequence and enhancing the expression of CamR can effectively prevent the leakage of the operon, and reducing the screening step length to 1bp may be able to screen out the CamR insertion site with higher induction efficiency.

           Moreover, there is significant room for improvement in the construction of our yeast cell factory. In the future, we plan to optimize the enzyme ratios in the metabolic pathway to achieve higher metabolic flux, enhancing substrate utilization efficiency and increasing final product yield, thereby approaching the detection range of biosensors.

           Additionally, the directed evolution system based on EvolvR still faces some challenging issues, such as undetectable off-target effects, limited mutability despite strong specificity, and difficulties in expressing and correctly folding due to large protein molecules. However, we have reason to believe that the continuous directed evolution system still holds unparalleled prospects. In the near future, many of these problems are expected to be effectively resolved. However, due to time constraints, we have not yet successfully tested the effectiveness of the continuous directed evolution system, nyEvolvR, in the production chassis. Moving forward, we aim to further evaluate the effect of the directed evolution system for specific synthetic enzymes and compare it with rational design, ultimately establishing an unconventional yeast platform that integrates production, endogenous detection, and directed evolution.

           During the exploring process, we encountered various problems, which were properly resolved through our continuous adjustments. For example, exploring fermentation conditions required repetitive trials, wasting a great amount of time and cost. We also explored the conditions for the efficiency of yeast transformation and methods of colony PCR validation in order to effectively screen positive clones. However, there are still some issues that we have not yet solved, such as the complexity of the EcN genome composition, making it difficult to obtain fragments in expected length via PCR, and the low efficiency of yeast co-transformation. Through the research and exploration process, we have learnt many things and broaden our horizon. We will continue to seek advice and keep exploring and devote our humble efforts to directed evolution.