In vitro characterization of the reporter
We linked mRFP downstream of the lactate, pH, and hypoxia-sensing promoters to construct a reporter, and determined the promoter response to lactate, pH, and oxygen by detecting the fluorescence of RFP. In the figure (A-C), the normalized fluorescence intensity of the lactate (pLldR), pH (pCadc) and hypoxia (pPepT) induced reporters are shown respectively, which we can draw the conclusion that reporter can respond to signals. To improve signaling stability as well as accuracy, we added amplifying genetic switches based on serine integrase (Bxb1, TP901) to the reporter to construct the AR strain (amplifying reporter). Figure (D-E) indicates lactate, pH and hypoxia induced AR with normalized fluorescence intensity.
Comparing figure A and D, it can be seen that the fluorescence intensity of the AR decreased significantly at a lactate concentration of 0 mM, and its expression was more stable over time. The fluorescence intensity of the AR was also greater at other concentrations of lactate induction, and the difference between the fluorescence intensity after lactate induction at each concentration was more pronounced. In contrast to Figure B and E, the fluorescence intensity of the AR appeared more stable over time at pH 7.3 and was higher than that of the reporter at pH 5.8, 6.3, and 7.3. Comparing figure C and F, the fluorescence expression of the AR was significantly higher after the addition of Switch under anaerobic conditions. This result indicates that the addition of amplifying genetic switch enhances the reporter intensity and robustness of three biosensors.
Meanwhile, we observed the mRFP fluorescence intensity of control group EcN and amplifying reporter strain AR after 48h of induction using a fluorescence microscope (Olympus BX53).
The results showed that the three promoters (pLldR, pCadC and pPepT)-Switch (TP901)-mRFP exhibited a uniform and clear red fluorescence signal after induction, indicating that the pLldR/pCadC/PepT-Switch (TP901) system could be expressed normally.
Testing for Cracker Reporter (CR strains)
We constructed the cracker reporter CR strains by adding the lysis gene phiX174E to the lactate, pH, and hypoxia-sensing promoters and attaching the amplification gene switch and mRFP to it. Figure (A-C) indicates the lactate, pH and hypoxia inducing reporters after the addition of the lysis gene phiX174E in induced and non-induced 's OD value. The lower $OD_{600}$ values indicate better lysis of the bacteria. Figure A, it can be seen that the $OD_{600}$ value becomes lower with increasing lactic acid concentration, and the OD value tends to a more stable state after 20 hours. Figure B, as the pH decreases, the $OD_{600}$ value also decreases. Figure C, the $OD_{600}$ under anoxic conditions was lower than that under normoxic conditions, indicating that our CR strain can respond well to the tumor environment.
Figure (D-F) indicates the fluorescence intensity of lactate (pLldR), pH (pCadc) and hypoxia (pPepT) induced reporters under induced and non-induced conditions after the addition of lysis gene phiX174E. Figure D, the fluorescence intensity showed an increasing trend with increasing lactate concentration. Figure E, the fluorescence intensity showed an upward trend with decreasing pH, and Figure F, the fluorescence intensity under normoxic conditions was very low, while the fluorescence intensity under hypoxic conditions increased significantly after 8h.
Figure (G-I) are the $OD_{600}$ of wild-type E.coli Nissle 1917 under induced and non-induced conditions which acted as control grop, and the wild-type bacteria could hardly respond to the induction of lactic acid, pH, and anoxic environments.
The results show that CR undergoes lysis under induced conditions, but the cells still produce fluorescence. It indicates that the fitted set of equations for lysis-growth should be a resonance function.
To further obtain the lysis-growth curve, we shortened the assay time to 30min a measurement. Figure A, changes in $OD_{600}$ of lactate-induced reporter under induced and non-induced conditions. Figure B, $OD_{600}$ changes of pH-induced reporter under induced and non-induced conditions. Figure C, change in $OD_{600}$ of hypoxia-induced reporter under induced and non-induced conditions. The results indicate that the lysis-growth curve is a dynamic function.
Next, we tested the constructed CR stains using CT26 cell cultures. In Figure A and D, CT26 cells were cultured for 5 consecutive days, after collecting the cell supernatant every 12 hours, the lactate concentration was measured. The $OD_{600}$ value and fluorescence response of the lactate-control CR strain was tested using this sample as the medium
In Figure A, as the lactate concentration in the culture increased, more bacteria were lysed and the $OD_{600}$ values decreased accordingly. In B and E, the pH level of the above cell culture medium sample was measured and used as the medium to test the $OD_{600}$ value and fluorescence response of the pH-controlled CR. Figure B, as the pH decreased, more bacteria were lysed and the $OD_{600}$ values showed a decreasing trend. Figure E, the fluorescence intensity shows an increasing trend as the pH decreases. Figures C and F, $OD_{600}$ values and fluorescence response of hypoxia-controlled CR tested using the above cell culture medium. Figure C, $OD_{600}$ values under anaerobic conditions were significantly smaller than those under normoxic conditions, and Figure F, fluorescence values under anaerobic conditions were higher than those under normoxic conditions. The results indicate that CR strains can respond in cell culture medium.
Utility Test for Cell Culture Samples
Lactate and pH induced promoter-controlled effector engineered strain co-incubated with RKO cells
Figure A shows the activity of RKO cells after incubation with each strain (OD=0.6, 30 μl, 3 hours) in fresh DMEM medium, normoxic conditions, and it can be seen that the relative viability of RKO cells in the experimental groups with the addition of effector strains in fresh culture medium, except for the plac+HlyE positive control, did not change significantly compared with the WT group.
Figure B shows the RKO cell activity of each strain after incubation in 3 day DMEM medium, normoxic conditions. It can be concluded that in the 3 day DMEM medium, due to the accumulation of metabolites such as cellular lactate, the lactate promoter and pH promoter were activated in the engineered strains and started to synthesize therapeutic proteins, resulting in a decrease in the relative viability of RKO compared to the WT group, especially in the pLldR+switch+HlyE and pCadC+switch+HlyE groups with the addition of the amplified gene switch.
Lactate induced promoter-controlled effector engineered strain co-incubated with CT26 cells
For the pLldR-Switch (TP901)-HlyE transformed effector, it was co-incubated with CT26 cells, and EcN was used as a control. CT26 cells after incubation with different bacteria were assayed using Calcein/PI Cell Activity and Cytotoxicity Assay Kit (C2015S, Biotime) with wild-type EcN as control and observed using fluorescence microscopy.
The principle of the kit is that two probes can detect intracellular esterase activity and cell membrane integrity, respectively, thus reflecting cell activity and cytotoxicity. Calcein AM stains live cells with green fluorescence, while Propidium Iodide (PI) stains dead cells with red fluorescence.
The results showed that the green fluorescence of cells stained by Calcein AM decreased significantly after co-incubation with the engineered bacteria, while the red fluorescence of cells stained by PI increased significantly. It was demonstrated that the cell activity was significantly reduced and the toxicity was significantly increased under the co-incubation of the engineered bacteria, i.e. the pLldR-Switch (TP901)-HlyE system had a significant effect on the treatment of tumor cells with CT26.
Hypoxia-inducible promoter (pPepT) control effector engineered strains were co-incubated with RKO cells.
Figure A shows the RKO cell activity after incubation with each strain (OD=0.6, 30 μl, 3 hours) in fresh DMEM medium, normoxic conditions. It can be seen that pPepT was not activated under normoxic conditions and the therapeutic protein was not expressed, so there was no significant change in RKO cell activity compared with the WT group; Figure B shows the RKO cell activity of each strain after incubation in fresh DMEM medium, anoxic conditions. Under hypoxic conditions pPepT was activated and the expression of therapeutic proteins caused RKO cells to be killed, so RKO cell activity was significantly reduced compared to the WT group.
Coincubation of different doses of effector engineered strains (OD=0.6) with RKO cells
Figure A shows the RKO cell activity after incubation with different doses of pLldR and pCadC control effector strains under 3 day DMEM medium, normoxic conditions. The RKO cell activity decreased with increasing doses of effector strains. Figure B shows the RKO cell activity after incubation with different doses of hypoxia-controlled effector strains under fresh DMEM medium, anoxic conditions. It can be seen that RKO cell activity decreased with increasing doses of effector strains controlled by hypoxic induced promoters.
30 μl effector engineered strains (OD=0.6) were co-incubated with RKO cells for different times
Figure A shows the RKO cell activity after incubation of plldR and pCadC control effector strains for different times under 3 day DMEM medium, normoxic conditions. It can be seen that the RKO cell activity decreased with the increase of co-incubation time. From the figure B, it can be obtained that RKO cell activity decreased with increasing incubation time of the effector strains controlled by pPepT under fresh DMEM medium, anoxic conditions.
Western blot
To verify the extracellular secretion of YebF-HlyE, we fused his tag at the C-terminus of YebF-HlyE. The expression of YebF-HlyE was detected by western blot. The results showed that the constitutive promoter could secrete HlyE under both inducible and non-inducible conditions as positive control, while the lactate (pLldR), pH (pCadc) and hypoxia (pPepT) inducible reporters could only secrete HlyE under inducible conditions and not under non-inducible conditions, which indicated that our engineered microbe could well cope with environmental induction and secrete HlyE in the tumor microenvironment.
Adhesion Analysis
To make the engineered strain are able to better adhere to colorectal cancer cells, we used engineered bacteria carrying the adhesion proteins INP-HlpA and mRFP, and the experimental group compared to the control group had red fluorescent spots all around the DAPI-stained cells, indicating that the HlpA could effectively improve the adhesion of EcN to colorectal cells. (where CT26 and RKO were magnified 400-fold, while RKO was magnified 300-fold), and this change was particularly evident in RKO cells.
We also added quantitative experiments after the advice of ONCE-CHINA in the meeting up. We co-cultured the RKO cell and bacteria carrying the adhesion proteins INP-HlpA , the control group is bacteria without INP-HlpA. After co-culturing, we used PBS to wash down the rest bacteria. As we can see in this chart, control group without adhesion protein was washed down more bacteria than experiment group, which indicates our adhesion protein did have effect. By the way, the number of colonies in the experimental and control groups before they were not co-cultured with cells were 38 and 47 respective.
