Overview
We attempted to construct kill switches in both Saccharomyces cerevisiae and E. coli and achieved some success. In Saccharomyces cerevisiae kill switch construction, we utilized an epigenetic modification system to suppress the toxic gene and replaced different promoters, using the sfGFP reporter plasmid to assess the gene suppression efficiency. In E. coli kill switch construction, we employed epigenetic modifications and CRISPRi technology to suppress the toxic gene, used the RFP reporter plasmid to assess gene suppression efficiency, and continually reduced gene leaky expression. This page will present experimental details of our construction of kill switches in Saccharomyces cerevisiae and E. coli.
Create epigenetic modification tool through the fusion of dCas9 and Sir2
Using computational tool, we designed primers to obtain dcas9 and sir2 deacetylase enzymes through PCR. These enzymes were subsequently linked together using a linker sequence. Additionally, we designed gRNA sequences targeting GFP to construct a tool plasmid. This tool plasmid was then introduced into the Saccharomyces cerevisiae strain BY4742. Subsequently, we observed the fluorescence intensity under a microscope.
Figure 1. Comparison of fluorescent effects. a. Experimental group with tool plasmids exposed to standard lighting conditions. b. Control group exposed to standard lighting. c. Experimental group with tool plasmids excited for fluorescence. d. Control group excited for fluorescence.
Under the microscope, it was observed that, compared to the control group, the strains containing the engineered plasmid exhibited a certain reduction in fluorescence, providing evidence that the deacetylase significantly suppressed the expression of the GFP gene to a certain extent.
After inserting the MazF or EcoRI, controlled by constitutive promoter, into the pRS415 vector, we constructed two different toxic gene plasmids, pR_DK02 and pR_DK03, and subsequently introduced them into Saccharomyces cerevisiae BY4742 for toxicity testing.
Figure 2. Validation of toxic gene effects. a. Transferring empty vector pRS415 into Saccharomyces cerevisiae BY4742 resulted in abundant colony growth. b. Transferring toxic gene EcoRⅠ into Saccharomyces cerevisiae BY4742 resulted in sterile colony growth.
From the figure, it is evident that the toxic genes exhibit strong toxicity in Saccharomyces cerevisiae, which is advantageous for the subsequent safety of the kill switch.
We successfully obtained dcas9 and Sir2 deacetylase enzymes through PCR and linked them together using a linker sequence. This fusion protein was then connected to a vector, resulting in the successful construction of the tool plasmid pR_DK01. Subsequently, pR_DK01 was introduced into the Saccharomyces cerevisiae strain BY4742, and colony PCR was performed on the transformed cells to confirm the presence of correctly validated strains.
Figure 3. The strains with pR_DK01 can grow normally.
Subsequently, we divided each toxic gene plasmid into two fragments through PCR and simultaneously introduced both fragments of the same gene into the previously validated strain mentioned above. The aim was to delay the expression of the toxic gene, allowing ample time for the engineered plasmid to carry out targeted inhibition.
Figure 4. The growth status of strains containing the kill switch. a. Transferred the pR_DK03 plasmid into Saccharomyces cerevisiae containing pR_DK01. b. Transferred the empty vector pRS415 into Saccharomyces cerevisiae containing pR_DK01 as a control.
We observed a significant reduction in the number of colonies containing the toxic gene compared to the control group after transformation. To validate the correct transformants, we conducted plasmid loss experiments. During the plasmid loss process, we noticed that the colonies containing the toxic gene did not yield PCR bands, suggesting the strain's instability. We hypothesized that this instability might be due to the continued strong expression of the toxic gene. As a result, we plan to explore alternative and more potent epigenetic suppression methods in our subsequent work to achieve successful construction of the kill switch.
Enhance epigenetic modification elements by incorporating specific sequence
We designed primers using a computational tool and performed PCR to obtain MazF and EcoRⅠ genes with homologous arms from the toxic gene plasmids. Subsequently, we obtained HML-I and HML-E silencer from the Saccharomyces cerevisiae genome and successfully constructed plasmids pR_DK04 and pR_DK05 by Gibson assembly, combining the toxic genes with HML-I and HML-E silencer, and the vector. These plasmids were then separately introduced into Saccharomyces cerevisiae BY4742, resulting in the successful transformation of the strains.
Figure 5. Strain growth status containing HML-I and HML-E silencer along with a kill switch. a. Transferring pR_DK04 into Saccharomyces cerevisiae strain BY4742 resulted in a great number of transformants. b. Transferring pR_DK05 into Saccharomyces cerevisiae strain BY4742 resulted in a great number of transformants.
Subsequently, we conducted colony PCR and sequencing on multiple transformants, obtaining several strains with correct sequences. This confirmed the successful suppression of the toxic genes using the proposed method.
Figure 6. Colony PCR verification to validate the above-mentioned transformants. a. Colony PCR verification to validate the pR_DK05. b. Colony PCR verification to validate the pR_DK04.
Testing tool components by replacing the promoter
We used sfGFP as a reporter element to measure the suppression efficiency and employed two different promoters: the CHO promoter and the CYC1 promoter. We measured their fluorescence values before and after the tool element's functionality. We observed a significant decrease in fluorescence values, indicating that the tool plasmid can effectively suppress gene expression. Furthermore, when using the CYC1 promoter, which is weaker compared to CHO, we observed a more pronounced reduction in fluorescence values. This suggests that switching promoters can enhance the degree of gene suppression by the tool plasmid.
Figure 7. Suppression of the sfGFP gene with CHO promoter and CYC1 promoter by the tool components
Inhibiting gene expression through Dam methylation
We designed primers using a computational tool and employed synthesized primers from Beijing Tsingke Biotech company for annealing. This allowed us to obtain fragments of the mioC and dnaAP2 promoters. The results are illustrated in the accompanying figure.
Figure 8. Verification of mioC (a) and dnaAP2 (b)
Using seamless cloning, we linked the mioC and dnaAP2 promoters upstream of the RFP fluorescent protein, resulting in the plasmids pR_DK06 and pR_DK07. We conducted colony PCR analysis to confirm the connection of mioC and dnaAP2 to the upstream region of RFP.
Figure 9. Growth status of E. coli after transformation. a. pR_DK06 transferred to E. coli. b. pR_DK07 transferred to E. coli.
Figure 10. Verification of pR_DK06. a. pR_DK06 upstream. b. pR_DK06 downstream
We separately transformed the extracted pR_DK06 and pR_DK07 plasmids into E.coli strains MG1655 and JM110. The results are depicted in the figure
Figure 11. Growth status of plasmid transformed E. coli. a. pR_DK06 transferred to JM110. b. pR_DK06 transferred to MG1655. c. pR_DK07 transferred to JM110. d. pR_DK07 transferred to MG1655.
We transformed the pR_DK06 and pR_DK07 plasmids into two different E.coli strains, namely JM110, which lacks the endogenous Dam methyltransferase, and MG1655, which possesses the endogenous Dam methyltransferase.
It can be observed that the fluorescence intensity of JM110 transformants with pR_DK06 was higher than that of MG1655 transformants. However, the fluorescence intensity of JM110 transformants with pR_DK07 was lower than that of MG1655 transformants.
Figure 12. The growth status of MG1655 and JM110 strains containing the pR_DK06 plasmid or the pR_DK07 plasmid
Preliminary experiments indicate that the Dam methyltransferase can indeed inhibit downstream gene expression of the mioC promoter but does not exert inhibition on the dnaAP2 promoter. Therefore, mioC was selected for further investigation in subsequent experiments.
To further validate the inhibitory effect of methylation on the mioC promoter, we measured the OD values of strains containing the Dam methyltransferase (MG1655) and strains with Dam methyltransferase knocked out (JM110) after 12 hours of culture using an microplate reader. Additionally, we quantified the fluorescence intensity of the pR_DK06 plasmid expression. We also calculated the fluorescence intensity per unit bacterial concentration. The results indicated that the fluorescence intensity of pR_DK06 plasmid expression in MG1655 strains, where Dam methylation was present, was lower compared to the normal expression of pR_DK06 plasmid in JM110 strains. This preliminary experiment suggests that Dam methylation has an inhibitory effect on the mioC promoter.
Figure 13. Unit fluorescence values of JM110 and MG1655 containing pR_DK06
Enhancing gene repression by combining CRISPRi with Dam methylation
We transformed plasmids containing the ccdB gene into E. coli MG1655, using MG1655 with the plasmid pR_DK06 as a blank control. It was observed that the E. coli cells transformed with the ccdB gene exhibited minimal to no growth.
Figure 14. Verification of the ccdB toxic gene. a. MG1655 with ccdB. b. MG1655 with pR_DK06.
We employed Gibson assembly to ligate the sgRNA scaffold onto a plasmid regulated by the lacZ promoter that controls dCas9 expression. Subsequently, we validated this construction through primer design.
Figure 15. Validation of five different tool plasmids through PCR
After selecting single colonies with the correct band size, we extracted the plasmids and conducted sequencing. This process yielded correctly sequenced tool plasmids. Subsequently, we utilized Golden Gate assembly to interchange different sgRNAs, resulting in the generation of five distinct tool plasmids.
Figure 16. MG1655 with pR_DK15 and pR_DK06.
The mioC-1 tool plasmid (pR_DK15) was introduced into MG1655 and selected on a culture medium containing chloramphenicol. After obtaining the correct transformants, they were rendered competent under 1000X induction. Subsequently, the pR_DK06 plasmid was introduced, and selection was carried out on a culture medium containing ampicillin and chloramphenicol. The correct transformants were then cultured in liquid medium for 18 hours, and the fluorescence intensity was measured using a microplate reader.
Figure 17. Fluorescence values measured under different conditions.
Enhancing gene repression using the fusion protein of dCas9 and Dam
We employed the Gibson assembly technique to construct the fusion protein of dCas9 and Dam. Following transformation into E.coli, selection was carried out on a culture medium containing chloramphenicol. Colony PCR was performed using designed primers, and single colonies with the correct band size were selected for liquid culture. Plasmid extraction and subsequent sequencing were conducted, resulting in the successful acquisition of the pR_DK08 plasmid.
Figure 18. Verification of dCas9 and Dam fusion protein plasmid
Design diagnostic test paper
The fluorescence changes within 24 hours after induction were observed on test strips under the following conditions:
1.IPTG: 1.4 mg/L (1000x), 2.8 mg/L (500x), 14 mg/L (100x), 70 mg/L (20x).
Figure 19. Test paper result chart of IPTG-RFP
2.Doxycycline (fourfold serial dilutions): 2 mg/L (1000x), 4 mg/L (500x), 20 mg/L (100x), 100 mg/L (20x).
Figure 20. Test paper result chart of tet-RFP
The fluorescence intensity in the detection zone of the test strip reached its peak between 8 to 12 hours after contact with the sample. Subsequently, it gradually increased with time, albeit with a slight decrease in fluorescence intensity. Within a certain range, as the concentration of the inducer increased, the brightness of the fluorescence also increased.
Figure 21. The fluorescence effects in cleanroom and non-cleanroom environments.
After 24 hours, the gel was washed, and the wash solution was streaked onto solid culture media containing the corresponding resistance markers and left to incubate overnight. It was observed that the cells remained viable, indicating the possibility of strain leakage. This suggests the need for the application of a kill switch.
Figure 22. 24h scribbling of Washing liquid.
