Biotechnology applications require engineering complex phenotypes, and thus, a complete reprogramming of innate metabolism. However, the lack of knowledge of the genetic basis of complex traits restricts our ability to modify genes sequentially, and the size of the combinatorial mutational space spanning complex phenotypes is much larger than the size that can be searched on laboratory timescales. Therefore, engineering complex phenotypes at the systems level is a more feasible approach. To this end, we developed the novel CRISPR–Cas nuclease and CRISPR-enabled trackable genome engineering methods, which can be used as universal tools for genome-wide scale reprogramming of innate metabolism.

Given the modular structure of such enzymes, we hypothesized that engineering chimeric sequences would generate non-natural variants that span the kinetic parameter landscape, and thus provide for the rapid selection of nucleases fit for a particular editing system. Here, we design a chimeric Cas12a-type library with approximately 560 synthetic chimeras, and select several functional variants. We demonstrate that certain nuclease domains can be recombined across distantly related nuclease templates to produce variants that function in bacteria, yeast, and human cell lines. We further characterize selected chimeric nucleases and find that they have different PAM preferences and the M44 chimera has higher specificity relative to WT sequences. This demonstration opens up the possibility of generating nuclease sequences with implications across biotechnology.

We developed CRISPR-enabled trackable genome engineering methods, which reported that CRISPR–Cas gene editing in combination with massively parallel oligomer synthesis can enable trackable editing on a genome-wide scale. Then, we applied the methods to site saturation mutagenesis for protein engineering, gene expression regulation for pathway engineering, and perturbing the regulatory network for complex phenotypes. As one example of an application in E. coli, we designed and constructed a regulatory network library consisting of a total of 110,120 specific mutations in 82 regulators. There are ~4000 genes that interact in this regulatory network. We then used this library for different targeted phenotypes that are important for industrial applications, and identified a series of positive mutations in a single round of selection which improved these traits, including overcoming inhibition of cell growth (furfural), increasing organic solvent tolerance (styrene), developing low carbon feedstocks (acetate), and improving biofuel resistance and production (isopropanol and isobutanol).

As another application in yeast, we constructed a large scale of saturation mutagenesis library targeting on the activity sites of transcriptional regulators those interacted with more than half of yeast genes. Then, we screened and selected for C3-C4 alcohol tolerance without modifying the producing pathway. We found a series of isopropanol and isobutanol tolerant strains. The WAR1_K110N variant has resistance to both isopropanol and isobutanol. In addition, we identified that the genes related to glycolysis is important and specific to isobutanol tolerance. Our findings provide insights into the cellular response of yeast to C3-C4 alcohol, and mechanisms underlying specific tolerance to isopropanol and isobutanol. Without modification of metabolic pathway, the verified tolerant strains can be used as robust yeast strains for C3-C4 alcohol production.

References

1. Rongming Liu; Liya Liang; Emily F. Freed; Alaksh Choudhury; Carrie A. Eckert; Ryan T. Gill*. Engineering regulatory networks for complex phenotypes in coli. Nature Communications, 2020, 11, 4050.
2. Rongming Liu*; Liya Liang; Emily F. Freed; Ryan T. Gill*. Directed evolution of CRISPR/Cas systems for precise gene editing. Trends in Biotechnology, 2020, 39(3): 262-273.
3. Rongming Liu; Liya Liang; Emily Freed; Hao Chang; Eun Joong Oh; Zeyu Liu; Andrew Garst; Carrie A. Eckert; Ryan T. Gill*. Synthetic chimeric nucleases function for efficient genome editing. Nature Communications, 2019, 10, 5524.
4. Rongming Liu; Liya Liang; Andrew D. Garst; Alaksh Choudhury; Violeta Sànchez i Nogué; Gregg T. Beckham; Ryan T. Gill*. Directed combinatorial mutagenesis of Escherichia colifor complex phenotype engineering. Metabolic Engineering, 2018, 47: 10-20.
5. Rongming Liu; Liya Liang; Alaksh Choudhury; Marcelo C. Bassalo; Andrew D. Garst; Katia Tarasava; Ryan T. Gill*. Iterative genome editing of Escherichia colifor 3-hydroxypropionic acid production. Metabolic Engineering, 2018, 47: 303-313.
6. Rongming Liu; Liya Liang; Alaksh Choudhury; Andrew D. Garst; Carrie A. Eckert; Eun Joong Oh; James Winkler; Ryan T. Gill*. Multiplex navigation of global regulatory networks (MINR) in yeast for improved ethanol tolerance and production. Metabolic Engineering, 2018, 51: 50-58.
7. Liya Liang, Rongming Liu, Emily F. Freed, Carrie A. Eckert, and Ryan T. Gill*, Transcriptional regulatory networks involved in C3-C4 alcohol stress response and tolerance in yeast, ACS Synthetic Biology, 2021.01, 10, 19-28.