According to the on-target and off-target sequences determined by the dry experiments, we introduced the on-target and off-target into the EP1 and P15A plasmid segments respectively through PCR, and then verified the correctness of the PCR amplification bands through DNA agarose gel electrophoresis (Figure 1). Subsequently, we purified the PCR product and removed impurities such as enzymes using Dpn Ⅰ, and linked the plasmid backbone to on-target and off-target through gibson assembly. The other screening genes contained in the plasmid have been linked to the plasmid backbone using the same method as in the previous experiments.
We obtained Escherichia coli (E. coli) with plasmids through electroporation, coating, and spreaking. Finally, through inoculation, plasmid extraction, and DNA sequencing, we obtained the expected designed EP1-on, EP7-wt, and P15A-off plasmids (Figure 2).
After plasmid construction, we validated the screening function of the three plasmid system using MAD7 (WT).
EP1-on validation: Three plasmid systems were electroporated and introduced respectively into two E. coli K-12 competent cells containing EP1 plasmid. After recovery, the cells were inoculated onto chloramphenicol (Chl) + Ampicilin (Amp) + 0.1%Sucrose, and Chl + Amp culture media using drop inoculation method, respectively.
P15A-off validation: the three plasmid systems were electroporated and introduced into two E. coli K-12 competent cells containing EP1. After recovery, they will be inoculatd onto Chl + Amp + 0.1%Sucrose and Chl + 0.1%Sucrose culture media using the dilution drop method respectively.
Due to the lethal effects of sacB and Amp, the CFU grown normally in the experimental group was significantly different from that in the control group. The results of DNA sequencing and validation experiments indicated that we have successfully constructed a three plasmid system containing EP1-on, EP7-wt, and P15A-off, and the system performed its screening function normally (Figure 3).
The levansucrase expressed by sacB gene catalyzes sucrose hydrolysis and levan synthesis, leading to the death of E. coli cells. Based on the different toxicity of sacB mutations predicted by the dry experiments, we designed primers for two mutations in the sacB gene, S164A and E262Q, and constructed EP1-on-sacB_S164A and EP1-on-sacB_E262Q plasmids. Through electroporation, expansion culture, plasmid extraction, and DNA sequencing, we successfully obtained two types of EP1-on plasmids carrying sacB_S164A and sacB_E262Q mutants, respectively.
Finally, we successfully tested the lethality of sacB_S164A and sacB_E262Q mutants in E. coli. We transformed the aforementioned two EP1-on plasmids into competent E. coli K-12 cells, and dropped onto Kanamycin (Kan) and Kan + 0.1% Sucrose media, respectively after recovery, with sacB_WT and sacB_S164T serving as the controls. The fatality rates of sacB_S164A and sacB_E262Q were 19.33% and 41.35%, respectively. (Figure 4)
The yield of levan synthesized by levansucrase is affected by sucrose concentration, and thus the fatality rate of sacB is affected. In order to explore the rule between sucrose concentrations and the fatality rates of sacB, sucrose concentration gradients of 0.01%, 0.1% and 1% were set to detect the fatality rate of sacB mutants. According to the results, we found that the fatality rate of sacB increased significantly with the increase of sucrose concentration. Through the combination of sacB mutants and sucrose concentrations, we obtained sacB screening kits covering different fatality rates.(Figure 5-7)
We chose to use error-prone PCR to introduce random mutations into the MAD7 gene. We designed the preliminary experiments to verify the optimal conditions for error-prone PCR under medium and high mutation rates. We designed two different sets of primers (referred to as Primer 1/2) and established a temperature gradient of 54-61℃ based on primer properties. Under the aforementioned conditions, DNA agarose gel electrophoresis was performed on the MAD7 mutant genes amplified by error-prone PCR. Based on the brightness of the correct electrophoretic bands and impurity bands, we determined the conditions for error-prone PCR as: Primer 1, annealing temperature at 61℃, and extension time for 4 minutes (Figure 8).
After obtaining the MAD7 mutation library, we need to ligate the MAD7 mutant fragments with the EP7 plasmid. First, using the EP7-wt plasmid as a template, we obtained a large number of backbone fragments through PCR amplification (Table 1). Subsequently, after gel electrophoresis, Dpn I digestion, etc., the backbone fragments were extracted and purified. Then, we ligated the MAD7 mutation fragments with the backbone through gibson assembly. Finally, through electroporation, plating, scraping, plasmid extraction, etc., we collected all EP7-mutant plasmids.
We introduced the EP7-mutant plasmid into E. coli K-12 competent cells containing EP1-on and P15A-off via electroporation, establishing a three plasmid screening system. After recovery, the E. coli cells were dropped onto a medium containing Chl + Amp+ 1% Sucrose, and also onto a Chl medium as a negative control. MAD7(WT) was used as a positive control (Figure 9). Finally, we obtained 14 and 24 single colonies at concentrations of 10-2 and 10-1 using the drop inoculation method. The off-target rate of the MAD7 mutant included may be significantly lower than that of MAD7 (WT).
After spreaking, 7 out of the 38 single colonies selected above could not grow on Chl + Amp medium, indicating that the MAD7 mutant had undergone off-target cleavage. Therefore, we eliminated these 7 mutants.
The E. coli strain obtained through screening still contained two plasmids, EP7-mutant and P15A-off. We used serial passaging and gel extraction methods to remove the P15A plasmid from the E. coli cells, obtaining the EP7 single plasmid containing the mutant fragment. To calculate the editing efficiency of each MAD7 mutant, E. coli K-12 cells containing EP7-mutants and EP1-on plasmids were dropped onto 1%Sucrose + Chl and Chlmedia, with MAD7(WT) serving as a control. To calculate the off-target rate of each MAD7 mutant, E. coli K-12 cells containing EP7-mutants and P15A-off plasmids were dropped onto Amp + Chl and Chl media.
After calculating the editing efficiency and off-target rates (Figure 10, 11), we discovered several mutants in the high-precision MAD7 mutants obtained through screening that can significantly reduce the off-target rate while ensuring the high editing efficiency. Among them, the mutant No. 6 has an off-target rate of 3.17% when both on-target and off-target existed. Compared with the off-target rate of 99.02% for MAD7(WT), the off-target rate of the mutant No. 6 has decreased by 96.8%, and the editing efficiency is 59.51% of MAD7(WT). This mutant MAD7 is named HF-MAD7. HF-MAD7, with its low off-target rate and stable editing efficiency when editing the HBsAg gene target, may become a reliable tool for gene editing therapy of hepatitis B.
