Team:HZAU-China/Proof Of Concept
Proof of Concept
The goal of our project is to repair damaged enamel with engineered Escherichia coli Nissle 1917 in ultimate implementation, however, we demonstrated the feasibility of our project in E. coli DH5 α and BL21 (DE3) . As described on the Design page, our project contains three core function modules and a suicide module. Three core modules were connected by a simple "Exclusive-OR gate" signal processing circuit thus the engineered bacteria can respond differently to different teeth conditions and restore the use' s enamel efficiently. The final product will be a "Brace", along with some other hardware and software for the convenience of users.
To proof our concept, we have conducted a lot of experimental works in about two months after returning to Wuhan. Most core components of our project have been successfully demonstrated. However, due to time constraints, some of our experimental works are still in progress. However, through the preliminary experimental results, we determine that our project can be verified, and it has the potential to be developed into a complete product in the future.
Demonstration of engineered bacteria
The primary experiments were implemented in common E. coli strain DH5α or BL21(DE3). "Detection & Report" module, "Repair" module and "Sterilization" module are the core functional modules of our engineered bacteria. "Detection & Report" module can receive and transfer CSP signal of Streptococcus mutans [1], and produce fluorescent proteins as reporter. The secondary signal generated by the "Detection & Report" module will determine whether to open the "Repair" module or the "Sterilization" module. The "Sterilization" module is responsible for producing protein that kill S. mutans and removing S. mutans that may exist on the surface of the tooth. After ensuring the non-existent of S. mutans, the "Repair" module will produce repair protein to reform enamel. Among the three core modules, the "Repair" module and "Sterilization" module are the most important function output modules, which determine the basic functions of engineered bacteria. Therefore, the "Sterilization" module and the "Repair" module are the key nodes of our project. Around these key nodes, we demonstrated our engineered bacteria through experiments and other methods.
The outline of our project is shown below:
Figure 1. The outline of genetic circuit.
"Detection & Report" module
In our project, the "Detection & Report" module can sense the amount of S. mutans by detecting the concentration of CSP. The promoters nlmC and nlmAB can sense the CSP through the ComDE two-component system [2-3]. These two promoters are expected to have different response thresholds to distinguish different CSP concentrations. Besides, when the CSP concentration is higher than threshold, the "Detection & Report" module will produce fluorescent protein as a report. The circuit of this part is shown in Figure 2.
Figure 2. Design of this module.
Whether the ComDE two-component system and two nlm promoters from Gram-positive S. mutans can work in E. coli is the main problem in this part. To verify that, mCherry was placed under the control of PnlmC. The plasmid was transferred to E. coli DH5α. After induced with sCSP, the value of RFU/OD600 of the bacteria sample added 1 mg/mL sCSP was obviously higher than the group without induction, although there were no significant differences between other bacteria samples. Because of the limited time of wiki freeze, we didn't conduct further experiments in time, we assumed that the threshold value of added sCSP was in between 0.5 mg/mL and 1 mg/mL.
Figure 3. The expressing condition of mCherry by activation of PnlmC after added different concentrations of sCSP.
Besides, in the "Detection & Report" module, we also hope that the two promoters can have different CSP response thresholds. The engineered bacteria can distinguish the concentrations of CSP and produce fluorescent proteins with different colors to report the conditions of teeth.
Based on the above conclusions, we can proof that the "Detection & Report" module is feasible. More detailed experiments are still in progress.
Exclusive-OR gate
As described on the Design page, engineered bacteria will have three different response patterns based on the concentration of CSP. The "Repair" module and "Sterilization" module will not open at the same time. Therefore, we regard the relevance of the two modules as an Exclusive-OR gate. To realize that, engineered bacteria need a signal processing circuit. After the "Detection & Report" module receives the CSP signal, T7 RNAP will enter the signal processing circuit as a secondary signal molecule to activate the "Sterilization" module. At the same time, TetR will be expressed to turn off the "Repair" module. (Figure 4).
Figure 4. Design of the singnal processing circuit.
To verify this part, phoA-LRAP-LAA was replace with mCherry, and phoA-clyR was deleted (Figure 5). Plasmid was transferred to E. coli BL21(DE3). Single colone was selected from the solid medium and cultured for 3 h at 37°C. After that, cultures were inoculated to fresh LB medium containing kanamycin (final concentration 12.5 mg/ mL) and IPTG (concentration gradient 50 μM, 25 μM, 10 μM, 0 μM), then cultured for 3.5 h at 37°C, centrifuge at 4°C at 5000 rpm for 1 min. As shown in Figure 6, with the increase of IPTG concentration, the red fluorescence of bacteria was reduced.
Figure 5. Demonstration circuit of the signal processing circuit.
Figure 6. Fluorescent change of E. coli treated with different IPTG concentrations.
Besides, we also compared the fluorescent intensity of IPTG induced group with the control group by Synergy H1 microplate reader, and a significant decrease was observed (Figure 7).
Figure 7. Fluorescent intensity of bacteria induced with different IPTG concentrations. The error bars indicate standard errors (SEM) of three independent biological replicates.
Through these results, we can verify that the Exclusive-OR gate can work. It can turn off the "Repair" module when the "Sterilization" module is working.
"Sterilization" module
The function of the "Sterilization" module is to produce and secrete ClyR protein to kill S. mutans on the teeth surface [4]. Whether this module can work directly determines whether the "Repair module" can repair enamel efficiently. Therefore, we spent more time to verify the functions of the "Sterilization" module.
By Ni2+-affinity chromatography, we have successfully obtained ClyR protein in E. coli BL21(DE3). To determine the dose-dependent and time-dependent lytic activity of ClyR, PBS resuspended S. mutans UA159 was adjusted to an OD600 of ~0.8. In the 96-well plates, 20 μL ClyR solution at different concentrations and 140 μL cell suspension were added in each well. The Synergy H1 microplate reader was used to measure the drop of OD600 at 37 ℃ for 1 h. The results obtained from triple independent experiments are shown below (Figure 8).
Figure 8. Dose-dependent and time-dependent lytic activity of ClyR. A. Variation curve of OD600 of S. mutans UA159 treated with different ClyR concentrations. B. Decrease in OD600 of S. mutans UA159 treated with different concentrations ClyR. The error bars indicate standard error (SEM) of three independent replications.
Besides, we also detected the remaining CFU of ClyR treated S. mutans. PBS resuspended S. mutans UA159 was treated with ClyR solution in 2 ml EP tubes for 10 min. At the end of the reaction, serial dilutions of each tube were plated on BHI agar and incubated at 37 ℃ overnight. The results obtain from triple independent experiments are shown in Figure 9.
Figure 9. Viable counts of ClyR treated S. mutans. The error bars are obtained from three independent experiments.
In our project, ClyR should be able to be secreted to the outside of the cell. Therefore, we fused ClyR with PhoA signal peptide. To demonstrate the secretion of PhoA fused ClyR, plasmid pET-28a(+)-PhoA-ClyR was transferred to E. coli BL21(DE3), same plasmid just without PhoA signal peptide as control. E. coli strains were cultured to OD600~0.6, induced with 0.2 mM IPTG and 1 % glycine, and then overnight cultured at 16 ℃. On the second day, take the supernatant and bacterial liquid for SDS-PAGE analysis. The results of SDS-PAGE indicated that the PhoA fused ClyR can be secreted to the medium effectively (Figure 10).
Figure 10. SDS-PAGE analysis of ClyR secretion. BL-PhoA-: Bacteria liquid of PhoA nagtive group; S-PhoA-: Supernatant of PhoA negative group; BL-PhoA+: Bacteria liquid of PhoA positive group; S-PhoA+: Supernatant of PhoA positive group;
As can be observed from the above charts, ClyR can be expressed and secreted in E .coli, and it has a high lysis activity to S. mutans. Therefore, we think the "Sterilization" module can work.
"Repair" module
The "Repair" module can produce enamel repair protein LRAP. LRAP is an extensively researched enamel repair protein [5-8]. It has been proven to help hydroxyapatite nucleate and repair damaged enamel. Due to time and equipment constraints, we could not obtain sufficient quality LRAP protein for final verification. However, we have successfully expressed LRAP in E. coli and have done preliminary experiments. Currently the experiment of this module is still in progress.
To obtain the LRAP protein, we constructed pET-28a(+)-LRAP and transferred it to E. coli BL21(DE3). By Tris-Tricine-SDS-PAGE analysis, we inferred that LRAP can be expressed in E. coli (Figure 11).
Figure 11. Tris-Tricine-SDS-PAGE analysis of ClyR expression.
Since LRAP protein was hard to be purified, we cultured E. coli BL21(DE3) strains with or without plasmid transfer, plasmid transferred strain was induced with IPTG and the same strain without IPTG induction as control. The harvested bacteria was broken by ultrasound, then centrifuged at 10000 rpm, 4 ℃, and the supernatant was took for characterization.
To determine the mineralization capacity of LRAP, artificial saliva (8.7 mM KCl, 0.6 mM MgCl2, 10 mM CaCl2, 4.6 mM K2HPO4, 2.7 mM KH2PO4) was prepared. The broken supernatant was added into the artificial saliva and the final protein concentration was 5 mg/mL. The solution was incubated at 37 ℃ for 24 h and the turbidity was observed (Figure 12). We can see that the IPTG+ group had less precipitation and higher turbidity, which we speculated was due to the nucleation of LRAP protein and ions in artificial saliva, which hindered the formation of precipitation.
Figure 12. Effects of different supernatants on turbidity of artificial saliva.
Besides, we also design experiment called "Toothtern-blot" to verify the existence and function of LRAP. The teeth of domestic pigs were broken into fragments about 5 mm long on each side, which were selected and stored in 75 % alcohol at room temperature.
The total protein concentration of the three supernatants was adjusted into 10 mg/mL. Three pieces of pig teeth were washed with PBS buffer to remove the surface's alcohol, then put in three 2 mL centrifuge tubes respectively. After that, three supernatants with uniform protein concentration were added respectively. Incubate at 37 ℃ for 20 h, and the tubes were upside down for several times every 5 h to eliminate the influence of protein deposition processes as much as possible.
The incubated pig teeth fragments were repeatedly washed with PBS buffer for 2-3 min to remove the protein that was not bound to the teeth. The rinsed pig teeth were incubated in plastic petri dishes with diameter of 10 cm, diluted with 50 mL 5 % skimmed milk at 1:10000 and hybridized with HE-Tag, Rabbit pAb (YEASEN Biotech, Shanghai, China) at room temperature for 2 h. Then washed them with TBST buffer 3 times for 10 min each. After that, diluted with 50 mL 5 % skimmed milk at 1:10000 and hybridized with peroxidase-conjugated Goat anti-rabbit IgG (H+L) (YEASEN Biotech,Shanghai, China) at room temperature for 2 h. Then washed them with TBST buffer 3 times for 10 min each. Use A Super ECL Detection Reagent ECL (YEASEN Biotech,Shanghai, China)for development operations.
The results showed that LRAP was able to bind to the tooth fragments and had bond more in the dentin (Figure 13).
Figure 13. “Toothtern-blot” analysis of different supernatants. A and B. The same result in different exposures. C. Picture of the teeth in reality.
Due to time and equipment constraints, we could not obtain sufficient quality LRAP protein for final verification. However, we have successfully expressed LRAP in E. coli and demonstrated its nucleation and binding activity preliminarily. Currently the experiment of this module is still in progress.
Suicide switch
The suicide switch is a module that we added based on HP's feedback. It can restrict engineered bacteria to the braces, thus improving the safety of the project. The main components of this module are toxin MazF and promoter fhuA1 [9-11]. PfhuA1 is sensitive to Fe2+. Fe2+ can bind to fur protein, making it an active repressor, thus shutting down the transcription of PfuhA1 (Figure 14). In actual implementation, Fe2+ will be added to the "Braces". When the engineered bacteria leave the Fe2+-rich environment, MazF will be produced to kill the engineered bacteria.
Figure 14. Circuit of this module.
To demonstrate the function of this module, we transferred the circuit described in Figure 14 to the E. coli DH5α. Overnight cultured bacteria as the seed culture, plating on the M9 solid medium, and filter paper soaked with FeSO4 was placed on the plates, filter paper soaked with distilled water as control. As Figure 15 shows, the number of colonies growing around the filter paper soaked with FeSO4 was significantly higher than the control group.
Figure 15. The results of disk diffusion experiment.
Besides, we also detected the growth conditions of the E. coli strain in different Fe2+ concentrations (Figure 16). From these results, we can indicated that Fe2+ have no significant influence to the E. coli strain at experimental concentrations, and the growth of strain containing suicide module was inhibited when Fe2+ are not presence. Therefore, we can verify that the suicide module can work.
Figure 16. Bacterial growth of E. coli strain in different conditions. A. Growth conditions of MazF- strain in different Fe2+concentrations; B. Growth conditions of MazF+ strain in different Fe2+ concentrations.
References
[1] Lemos J A, Palmer S R, Zeng L, et al. The biology of Streptococcus mutans[J]. Gram‐Positive Pathogens, 2019: 435-448.
[2] Liu T, Xue S, Cai W, et al. ComCED signal loop precisely regulates nlmC expression in Streptococcus mutans[J]. Annals of microbiology, 2014, 64(1): 31-38.
[3] van der Ploeg J R. Regulation of bacteriocin production in Streptococcus mutans by the quorum-sensing system required for development of genetic competence[J]. Journal of bacteriology, 2005, 187(12): 3980-3989.
[4] Xu J, Yang H, Bi Y, et al. Activity of the chimeric lysin ClyR against common Gram-positive oral microbes and its anticaries efficacy in rat models[J]. Viruses, 2018, 10(7): 380.
[5] Le Norcy E, Kwak S Y, Wiedemann-Bidlack F B, et al. Leucine-rich amelogenin peptides regulate mineralization in vitro[J]. Journal of dental research, 2011, 90(9): 1091-1097.
[6] Kwak S Y, Litman A, Margolis H C, et al. Biomimetic enamel regeneration mediated by leucine-rich amelogenin peptide[J]. Journal of dental research, 2017, 96(5): 524-530.
[7] Mukherjee K, Ruan Q, Moradian-Oldak J. Peptide-Mediated Biomimetic Regrowth of Human Enamel In Situ[M]//Odontogenesis. Humana Press, New York, NY, 2019: 129-138.
[8] Shafiei F, Hossein B G, Farajollahi M M, et al. Leucine‐rich amelogenin peptide (LRAP) as a surface primer for biomimetic remineralization of superficial enamel defects: an in vitro study[J]. Scanning, 2015, 37(3): 179-185.
[9] Zhang Y, Zhang J, Hoeflich K P, et al. MazF cleaves cellular mRNAs specifically at ACA to block protein synthesis in Escherichia coli[J]. Molecular cell, 2003, 12(4): 913-923.
[10] Engelberg-Kulka H, Hazan R, Amitai S. mazEF: a chromosomal toxin-antitoxin module that triggers programmed cell death in bacteria[J]. Journal of cell science, 2005, 118(19): 4327-4332.
[11] Guan L, Liu Q, Li C, et al. Development of a Fur-dependent and tightly regulated expression system in Escherichia colifor toxic protein synthesis[J]. BMC biotechnology, 2013, 13(1): 25.
