Engineering

The aim of this project is to modify Lactobacillus through genetic engineering methods to make it produce Nisin, thereby enhancing its inhibitory ability against S. mutans. One of the key prerequisites for optimizing Nisin production is to have a deep understanding of the growth kinetics of Lactobacillus and S. mutans. Determining whether their growth conditions will change with the variation of sugar concentration in the culture medium is crucial for our project. We speculate that within a certain range, the higher the concentration of MRS sugar, the more conducive it is to the growth of bacteria. To verify this hypothesis and establish reliable culture parameters for further experiment, this experiment includes the growth of Lactobacillus, S. mutans, and E. coli (as a control group) in MRS media with different glucose concentrations. By regularly measuring the OD of the culture medium, the growth curves of Lactobacillus, S. mutans, and E. coli were plotted and analyzed.

This experiment prepared MRS liquid culture media with glucose concentrations of 2% (20 g/L, standard formula), 3% (30 g/L), and 4% (40 g/L), respectively. Among them, 3% and 4% of the culture medium were prepared by adding additional glucose to the standard MRS base medium. In the first experiment, we used the regular method (121°C, 45 minutes) to sterilize the culture medium with high-pressure steam. However, we found that the color of the culture medium significantly darkened after sterilization, and when used for plate streaking culture, the growth of each colony was severely inhibited, which means slow growth and small bacterial colonies. According to papers review, this phenomenon is due to the high sugar content (20 g/L) of the MRS culture substrate. Carbohydrates undergo Maillard reaction during long-term high-temperature and high-pressure sterilization, which not only causes browning of the culture medium, but also may produce by-products that inhibit bacterial growth. Based on this, we optimized the sterilization plan: the sterilization time of MRS basic medium was shortened to 15 minutes, and after sterilization, glucose concentration solution was filtered and sterilized (using a 0.22 μm filter membrane) and added to accurately adjust to the target concentration. Glucose solution is prepared by using sterile deionized water and solid glucose.

We inoculated Lactobacillus, S. mutans, and E. coli glycerol strains stored at −80°C onto MRS agar plates with corresponding glucose concentrations (2%, 3%, 4%). Then, we placed the agar plates in a 37°C incubator for 20 hours to obtain single colonies. Subsequently, independent colonies with good morphology were selected from each plate and inoculated into liquid MRS medium with corresponding glucose concentration for pre-culture amplification. After 15 hours of shaking culture at 37°C and 200 rpm, we transferred the pre-culture to fresh liquid MRS medium containing different glucose concentrations at an initial inoculation volume of 0.1 OD600. This time point is marked as the beginning of the formal experimental culture.

All liquid cultures were treated to shaking culture in a shaker at 37°C and 200 rpm. We used a spectrophotometer to regularly measure the optical density of the culture at a wavelength of 600 nm to measure growth. During measurement, we used fresh sterile MRS medium of the corresponding concentration as a blank control. In the first experiment, we planned to use a set of MRS culture medium as the blank control group for the entire OD measurement, but later we found that doing so could easily lead to contamination of the blank control group by miscellaneous bacteria. Therefore, in the second experiment, we prepared a fresh blank control group before each OD measurement.

Within a certain range, the growth of S. mutans is affected by changes in sugar concentration, but Lactobacillus and E. coli are not sensitive to changes in sugar concentration. However, we also noticed that there was a certain deviation between the data of the 3% glucose concentration experimental group and expectations, and the specific reasons (such as osmotic pressure effects or accumulation of metabolic byproducts) need further exploration. In addition, there are certain fluctuations in the experimental data, which can clearly reflect the overall growth trend. However, the reliability of individual data points still needs to be improved in further experiments by increasing biological replicates and optimizing sampling and measurement processes to enhance accuracy.

During the cultivation process, we observed a typical feature of S. mutans: it can form a clear white precipitate at the bottom of the culture bottle during the initial stages of expansion and liquid culture. This phenomenon is significantly different from Lactobacilli and E. coli (which usually precipitate during the dead phase). The papers point out that this is due to the fact that S. mutans tends to grow in a chain-like pattern, and its cells naturally settle under the action of gravity, which is an important biological characteristic of the strain and a sign of its active growth, rather than contamination or abnormalities.

This program aims to modify Lactobacillus to produce nisin in order to inhibit the growth of S. mutans. That is, we initially need to try to find the growth conditions which benefit Lactobacillus but not S. mutans, so we chose MRS culture medium—most conducive to the growth of Lactobacillus—for exploration. Streptococcus mutans and Lactobacillus exist in the oral cavity in the form of biofilms and are necessary conditions for the formation of dental caries. Therefore, based on our research conditions—sugar concentration—we designed experiments to explore the biofilm formation in MRS culture medium with glucose concentrations of 2%, 3% and 4% respectively, and provide a data basis for our project.

Our group chose quantitative measurement of biofilms by crystal violet staining to detect the formation of biofilm in MRS broth under those three glucose concentrations—2%, 3% and 4%. For Lactobacillus and E. coli, we set up two groups to minimize random error, while we set up three groups in S. mutans as the expansion culture of S. mutans is not that ideal.

After 24 hours, we removed the planktonic cell samples, used 1×PBS, 0.1% CV, diH2O and 95% alcohol in turn for rinsing, dyeing and rinsing. Then, we put the 96-well plate into the plate reader to get readings. The bacterial solution reading was subtracted from the corresponding culture medium reading, and then averaged to obtain the required data, that is, the growth of biofilms of the three bacteria in different glucose concentrations.

According to the average value of each bacteria growing under all three glucose concentrations, we found that within a certain range, the growth of S. mutans biofilm is affected by changes in sugar concentration; the higher the sugar concentration, the greater the amount of biofilm it generates. However, Lactobacillus and E. coli are not sensitive to changes in sugar concentration.

Nisin is a ribosomally synthesized antimicrobial peptide belonging to the lantibiotic family. It is initially produced as pre-NisA and undergoes a series of post-translational modifications mediated by the enzymes NisB and NisC, which dehydrate and cyclize the peptide, respectively. The mature pre-NisA is subsequently exported by the ABC transporter NisT and finally activated through leader peptide cleavage by the serine protease NisP, releasing the mature and bioactive nisin. In Lactococcus lactis, where the nisin biosynthetic system is naturally encoded, the modification complex composed of NisA/B/C localizes at the cell poles—preferentially the old poles—while NisT is distributed circumferentially along the membrane. NisB acts as a polar “recruiter” to coordinate modification and export, ensuring that only fully modified pre-NisA is secreted. This spatial organization tightly couples biosynthesis, modification, and export, thereby preventing premature secretion of inactive intermediates.

However, this architecture poses challenges for heterologous nisin production in other hosts, such as E. coli. Expression of nisA alone often leads to intracellular accumulation of unmodified pre-NisA with limited antimicrobial activity. Co-expression with nisB and nisC can promote proper dehydration and cyclization, but secretion remains inefficient without nisT. The inclusion of nisP completes the maturation cascade, enabling leader peptide cleavage and full activation of nisin in the extracellular medium. Thus, a complete nisABCTP module is expected to yield the highest antimicrobial efficacy.

We are using these plasmid set (using backbone for E. coli): a. pET-21a(+)-nisA (AmpR): nisA: BamHI/XhoI and N-terminal marker for characterization; C-terminal 6xHis tag for purification and separation; b. pRSFDuet-1-nisB-nisP (KanR); nisB: NdeI/XhoI, nisP: NcoI/BamHI; c. pACYCDuet-1-nisC-nisT (CmR); nisC: NdeI/KpnI, nisT: NcoI/BamHI. This will be hosted in E. coli DH5α for molecular cloning and E. coli BL21(DE3) for plasmid expression.

For our cloning and verification, we assemble the three plasmids and then verify by gel electrophoresis and DNA sequencing. Then, we transform sequentially with tri-antibiotic selection (Amp + Kan + Cm), confirming plasmid compatibility.

To do expression and fractionation, we grew in LB + tri-antibiotics to OD600≈0.5 at 37 °C. Then, we induced with 0.1–0.3 mM IPTG since all three plasmids need IPTG to be induced.

We looked at the antimicrobial and biofilm assays compared to S. mutans. To do this, we collected cell pellet, periplasm/membrane fraction (for NisT), and culture supernatant for inhibition assays. We will send the engineered bacteria to the culturing group for testing.

Early constructs confirmed nis gene compatibility. Anticipate need for full nisABCTP pathway for strong supernatant activity.

This project builds a synthetic circuit that integrates the S. mutans quorum-sensing pathway (ComD/ComE) with a nisin expression cassette on a single plasmid.

CSP (from S. mutans) → ComD senses and autophosphorylates → phosphate transfers to ComE (ComE~P) → ComE~P activates PcomE_like → nisA is expressed → Nisin inhibits S. mutans.

In future work, we plan to experimentally verify the functionality of this CSP-responsive nisin expression plasmid. Validation will include co-culturing the engineered Lactobacillus strain with Streptococcus mutans to examine whether nisin production is specifically triggered by CSP signaling. Quantitative assays such as agar diffusion inhibition tests and qPCR for nisA transcription will be employed to assess antimicrobial activity and promoter responsiveness.

Due to current time and resource constraints, these functional tests will be carried out at a later stage when adequate laboratory capacity becomes available. The plasmid construction itself has been completed and verified by sequencing, and the upcoming experiments will focus on confirming its regulatory performance and potential application in in vitro oral microbial community models.

We gained valuable experience in modular cloning and promoter engineering. We learned how to integrate two independent gene cassettes—comDE and PcomE_like-nisA—within a single plasmid backbone while maintaining correct orientation and transcriptional insulation. During the process, we also deepened our understanding of the S. mutans quorum-sensing pathway and how its signaling components can be repurposed for targeted antimicrobial responses. Troubleshooting restriction site compatibility and optimizing RBS spacing further improved our experimental design skills and highlighted the importance of detailed sequence planning in synthetic biology.

We wanted to test the efficiency of nisin's bacterial inhibitory effect against Streptococcus mutans using a simple disk diffusion assay on BHI plates.

We prepared a 10 mL nisin stock solution with 900 IU/mL potency. Using this stock solution, we made 5 working solutions with concentrations ranging from 50–150 μg/mL in increments of 25. We soaked sterile paper disks in the working solutions for 3 hours.

Five BHI agar plates were inoculated with S. mutans, and immediately after inoculation a single disk was placed on each plate. Each disk had been pre-soaked in a different concentration of nisin (ranging from 50–150 IU) for 3 hours. Plates were then incubated for 24 hours before measuring zones of inhibition.

Little to no inhibition was detected at these concentrations. We concluded that the nisin concentration levels used were either too low to produce a strong antimicrobial effect, or that the biofilm formation of S. mutans led to our results. Our findings prompted us to try higher concentrations in the next cycle as well as test in biofilm & non-biofilm conditions.

We wanted to test higher concentrations of nisin and to test whether biofilm formation explained the last cycle's results. We hypothesized that higher concentrations would give better measurable inhibition.

We prepared a 10 mL nisin stock solution with 1000 IU/mL potency. Using the stock solution, we created six 2 mL working solutions with concentrations ranging from 150–300 μg/mL in increments of 25. We prepared 14 BHI and 14 BHI + 2% sucrose plates (28 total).

We soaked two sterile paper disks in each solution for 3 hours. 14 plates (7 BHI & 7 BHI + 2% Sucrose) were designated as biofilm plates; they were inoculated with S. mutans and incubated for 12 hours before adding disks. 14 plates (7 BHI & 7 BHI + 2% Sucrose) were designated as non-biofilm plates; they were inoculated with S. mutans and disks were added immediately. We applied the disks with varying concentrations of nisin (150–300 IU in increments of 25 IU) onto all plates, then let the plates incubate for 24 hours.

Biofilm plates showed no zone of inhibition (as expected). Non-biofilm plates showed inconsistent results: sometimes lower concentrations produced a zone of inhibition while higher ones did not. We believed that pre-soaking disks in solution might not deliver consistent dosing and planned to apply nisin directly to plates in the next cycle.

To improve consistency, we decided to pipette nisin solutions directly onto the disks on seeded plates instead of pre-soaking disks.

We prepared 14 BHI and 14 BHI + 2% Sucrose plates (28 total). 14 plates (7 BHI & 7 BHI + 2% Sucrose) were designated for biofilm plates and 14 plates (7 BHI & 7 BHI + 2% Sucrose) designated for non-biofilm plates. Upon trying to make our selected concentration range (150–300 IU) within a set volume of 20 μL for direct application onto disks, we discovered that the amount of nisin powder required to reach these concentrations was too high for DI water solubility, and the volumes needed (166–333 μL per plate) were impractical for disk diffusion assays.

Despite the challenge, we applied the large volumes (166–333 μL) of varying nisin concentrations directly onto the plates in hopes of seeing inhibition. Biofilm plates were inoculated with S. mutans and incubated for 12 hours before adding nisin solution. Non-biofilm plates were inoculated before adding the nisin solution immediately.

No zones of inhibition were observed under any of the conditions tested. Upon reviewing the specifications of our commercially extracted nisin, we discovered that it contained only 2% active nisin. Although the product was labeled at 900 IU/mg, we concluded that the actual concentration was too low to produce measurable inhibition in a disk diffusion assay. This prompted us to introduce a positive control using ampicillin to confirm that our assay setup was functioning properly and that the absence of inhibition zones was due to the nisin preparation, not experimental error.

After observing no clear inhibition from nisin in previous cycles, we needed to confirm that our disk diffusion setup itself was valid. We introduced ampicillin, a well-established antibiotic against gram-positive bacteria as a positive control.

We prepared 6 BHI + 2% sucroe plates total. There were 3 biofilm plates and 3 non-biofilm plates. 1 biofilm and 1 non-biofilm served as our control.

We added one ampicillin disk to our plate, one 3000 IU nisin disk, and one 300 IU nisin disk into each experimental plate, both biofilm and non-biofilm. The plates were then incubated overnight.

We saw that there was clear inhibition from the positive control, validating our assay. We will move onto doing recombinant nisin production using cation-exchange chromatography and SDS page.

The AalR regulator was selected as the L-aspartate (L-Asp) sensor to control the safety switch of the killswitch system. AalR is a LysR-type transcriptional regulator derived from Acinetobacter baylyi ADP1, where it activates genes involved in aspartate metabolism in the presence of L-Asp. Since neither E. coli nor Lactobacillus reuteri naturally expresses AalR, the team designed an expression construct using the pUPD3 backbone with the p32 constitutive promoter to continuously express AalR. Downstream of the coding sequence, a strong double terminator (BBa_B0015) was included to prevent transcriptional read-through. The system was intended so that AalR, once expressed, could respond to environmental L-Asp and activate MazE transcription when the inducer is present, ensuring survival only under safe conditions.

The AalR construct was assembled as p32–RBS (BBa_B0032) > aalR > BBa_B0015 using the pUPD3 backbone, which provides chloramphenicol resistance. The medium-strength RBS was selected to ensure adequate protein production without imposing metabolic stress.

Transformed E. coli colonies will be selected using chloramphenicol to confirm successful construct integration. Functional testing will involve verifying AalR expression and assessing whether L-Asp could activate its target promoter, enabling MazE expression in downstream experiments.

Through this cycle, the team confirmed that AalR expression in E. coli was feasible and non-toxic, setting the foundation for coupling AalR-regulated MazE expression. They also learned that environmental L-Asp concentrations and promoter sensitivity would be key parameters for fine-tuning induction. Future steps include testing dynamic response curves to varying L-Asp concentrations and optimizing AalR–promoter binding for predictable regulation.

MazF, the toxin component of the MazEF system, is an endoribonuclease that cleaves mRNA and halts protein synthesis, ultimately causing cell death. It was placed under the strong T7 promoter in the pET-21(+) vector for precise control. This setup leverages IPTG-induced T7 RNA polymerase expression in E. coli BL21(DE3). To minimize premature toxicity, a weaker RBS (BBa_B0033) was selected, ensuring controlled toxin translation. The design also anticipated future transfer to L. reuteri, where IPTG induction is not viable, and planned to substitute the T7 system with a constitutive promoter.

The MazF construct was assembled as T7 promoter–BBa_B0033 RBS–mazF–terminator in the pET-21(+) backbone (ampicillin resistance). The backbone's built-in LacI repressor ensured that MazF remained inactive until IPTG induction.

Colonies will be selected on ampicillin-containing plates. Controlled IPTG induction experiments are planned to determine the optimal concentration range for toxin activation that would effectively kill the cells under simulated oral conditions.

This cycle revealed how promoter and RBS strength balancing is essential to avoid premature host death. The team also recognized that the dynamic range of IPTG concentration directly affects kill efficiency. Future improvements will involve optimizing promoter strength and exploring alternate inducers or repressors compatible with L. reuteri physiology.

MazE, the antitoxin, neutralizes MazF by forming a protein–protein complex that inhibits its RNA-cleaving activity. Its transcription was designed to be regulated by AalR, ensuring that MazE is expressed only when L-Asp is present, under safe conditions. The pT7 vector (kanamycin resistance) was used for mazE expression because it lacks an RBS and terminator—allowing insertion of custom regulatory sequences. The BBa_B0034 RBS was chosen for moderate translation efficiency to maintain a balance between neutralizing MazF and preventing metabolic overload.

The construct was assembled as AalR-regulated promoter – BBa_B0034 RBS – mazE – terminator in the pT7 backbone. The AalR-controlled promoter ensures context-specific expression depending on L-Asp availability.

Colonies will be selected using kanamycin resistance. Experiments will focus on determining safe and lethal L-Asp concentration thresholds: identifying ranges where MazE expression keeps E. coli alive and conditions where reduced L-Asp leads to MazF-mediated killing.

This cycle demonstrated how environmental sensing (via AalR) could directly modulate toxin–antitoxin balance. The team learned that promoter strength, inducer concentration, and timing are critical for maintaining system stability. These results guided refinements in adjusting expression ratios and potential feedback control mechanisms for more precise regulation.

The final design aimed to integrate all three modules into a coordinated system where L-Asp and IPTG levels jointly determine the cell's fate. AalR activates MazE expression under safe conditions (presence of L-Asp), while IPTG induces MazF toxin expression.

All three constructs—pUPD3 (AalR), pT7 (MazE), and pET-21(+) (MazF)—were co-transformed into E. coli BL21(DE3), leveraging their distinct antibiotic resistance markers for selection.

Dual induction experiments are planned to test how IPTG and L-Asp concentrations interact. Viability assays will measure system stability and switching accuracy under different inducer combinations.

Integration will teach the team how small variations in expression strength can disrupt the balance between toxin and antitoxin. This insight might emphasize the need for fine-tuning promoter strength and RBS activity to prevent leaky toxicity or incomplete killing.