1. galK gene inactivation

Our project aims to improve the gene editing efficiency by constructing a purification system (strainer). The describe of the strainer method can be found in home page. Gene inactivation is a frequently-used step in the strain construction. Thus, we use it as a good start for our method design and construction. The galK gene is one of the popular targeting gene in E. coli because the cells with the inactivation of galK gene will be observed as white colonies on the MacConkey agar. Moreover, the wild-type control will be observed as red colonies on the MacConkey agar. To this end, we designed four gRNAs (in the EC21 to EC24 plasmids) targeting on the different position of the galK gene and related donor DNAs contained stop codon (TAA). In addition, each gRNA with its related donor DNA is constructed in the same editing plasmid, which is good for genome editing using gRNA libraries.

We used two E. coli strains (EC85 and EC88) with the plasmid harboring lambda red system and different CRISPR proteins (Fig. 1) as host strains to test our four editing plasmids. Then, we can compare the editing efficiencies using four editing plasmids with two different Cas protein (MAD7 and AsCas12a). The editing plasmid with lowest editing efficiency can be used for the design and construction of the third plasmid of strainer system.

2. Construction and validation of the “strainer”

The previous studies have demonstrated that the SOS response is the first DNA repair system described in E. coli. The transcription of genes such as lexA, sula, umuD and recA are regulated by the SOS response. Thus, we can use the promoters of these genes to start the transcription of gRNAs targeting on the plasmid harboring a counter-selection markers if there is a double-stranded DNA breaks (DSBs) on the genome. When the sacB plasmid is cured by CRISPR/Cas system, the strain with successful recombineering can survival in the media with sucrose (Fig. 2). The strain without DSBs, still has the plasmid harboring sacB gene, cannot survival in the media with the sucrose. This is why we can increase the overall editing efficiency.

Firstly we designed a second plasmid harboring sacB gene in the EC85 or EC88, and used Trc inducible promoter to control its expression. Then, we designed another gRNA that can be inserted in the EC21 plasmid. This gRNA target on the second plasmid harboring sacB gene, and use the promoter of genes such as lexA, sula, umuD and recA.

3. Protein modification

The lethal mechanism of the sacB toxicity gene is that the protein expressed by the sacB gene (levansucrase) can catalyze the hydrolysis of sucrose into glucose and fructose, and polymerize fructose into levan. The large accumulation of levan in E. coli will kill the cells. The S164 forms a hydrogen bond with the nucleophilic agent D86 and the 4-OH of the fructose group, and S164 is important to ensure the position stabilization of D86 (Fig. 3). If we design some mutation in S164 site, the mutant could reduce the hydrolysis rate, and then the toxicity of SacB protein will reduced.

The position of the carboxyl group of D86 is restricted by hydrogen bonding. We speculate that the S164T mutation with an additional -CH3 group would change the orientation of the-OH and would effectively form new hydrogen bonds. If hydrolysis rate reduce, the toxicity of SacB protein will reduced. Thus, we test S164T mutation using Molecular dynamics (MD) simulations. The results showed that This mutation breaks the delicate balance of the ternary catalytic amino acid with the ligand (Fig. 4). Therefore, we speculate that it will lead to reduced cytotoxicity. To this end, we sought to apply S164T mutation in the wet-lab experiment.

4. Insertion of the long gene fragment

The “strainer” method improved the overall editing efficiency by removing the unedited cells. We have previously validated the feasibility of improving editing efficiency by introducing the "strainer" method for screening in the single gene editing, but we sought the "strainer" to potentially serve as a universal method for cell factory construction. Therefore, we applied the "strainer" to a large fragment gene editing process to verify the universality of the system.

SS9 is a safe locus in the E.coli genome, where we can insert long gene fragment without strong effect on the growth of E.coli. Therefore, we chose this region to verify the "strainer" for long DNA fragment insertion experiment. We chose the CRISPR/Cas9 gene editing method as the control group and used CRISPR/Cas9 to construct the “strainer” method as the experimental group, because the CRISPR/Cas9 is also commonly used in the genome editing field. We introduced the isopropanol (IPA) production pathway (~5000 bp) into the strains using both methods to compare the efficiency of long gene fragment insertion.

5. Fermentation for IPA production

We constructed the EC61 strain by inserting the IPA production pathway into the genome in MG1655, and then we further modified the EC61 strain by introducing the FliA_R94K mutation that confers isopropanol resistance.

We used the "strainer" and the original CRISPR/Cas methods for the introduction of FliA_R94K mutations into the EC61 strain. Then we set EC61 as the control group and EC65 as the experimental group. Three shake flasks were set up in each group to culture and ferment as the parallel samples, two of which were sealed with parafilm and one with tinfoil.