Overview

The main goal of our project is to degrade TMA with the engineered bacterium E. coli Nissle 1917 for the treatment of cardiovascular diseases. Nissle 1917 was chosen as the chassis because it has been widely demonstrated to be a kind of intestinal probiotics that can colonize the gut without unduly disrupting homeostasis in the intestine [1]. We firstly demonstrated the feasibility of our project in E. coli DH5α and BL21 (DE3). As described on the Design page, our project contains three functional modules and a suicide module. The three functional modules are linked by a simple theophylline small molecule, leading to respective responses to different patient conditions and an effective degradation of TMA to reduce thrombosis. The final product will be a 'brace', along with some hardware and software, to facilitate patients' usage.

In order to prove our concept, we carried out a great deal of experimental work over a period of about three months during and after the summer vacation. Most of the core components of our project have been successfully demonstrated. Although some of our experimental work is still in progress due to time constraints, with preliminary experimental results, we are certain that our project can be validated and will most likely to provide the public with a novel treatment idea in the future.

Demonstration of engineered bacteria

The primary experiments were implemented in common E. coli strain DH5α or BL21(DE3). The most central module in our project is the degradation module, followed by theophylline induction, validation of Cre/LoxP recombination efficiency, alleviation and suicide modules. In the degradation module, we express tmd, dmd and formaldehyde dehydrogenase gene clusters to degrade trimethylamine and metabolise the toxic by-product formaldehyde. It is crucial to test the concentration of theophylline that can induce downstream functions, while the efficiency of Cre/LoxP recombination, which is related to how long it takes for our patients to switch functional templates after taking the drug, is also important. The alleviation module is responsible for expressing small molecules that contribute to physical and mental health. Therefore, the "Degradation" module is the key node of our project. Around these modules, we demonstrated our engineered bacteria through experiments and other methods.

The outline of our project is shown below:

Theophylline induction

In our project, we set up a theophylline-induced riboswitch to turn on the expression of downstream genes. To test the activity of the riboswitch, we chose to attach a red fluorescent protein to the riboswitch and measure the expression intensity of the fluorescent protein.

The curcuit of this part is shown below: (Figure 2).

From a literature review[2], we know that theophylline induces well at the chosen induction concentration of 2 mmol/L. In order to simulate the reaction conditions in intestine, we allowed the engineered bacteria to incubate at 37°C for 9h and measured the fluorescence expression intensity after induction using a theophylline concentration of 2mmol/L. The results of the experiment are shown below (Figure 3).

Analysis of the experimental data shows that the riboswitch expression intensity induced by theophylline under 37°C incubation is more than 30 times that of the control, which can initially indicate a better induction effect, but of course we can also see that there is a slight leakage of this riboswitch under normal conditions.

If our vision can become a reality at some point in the future, we should consider reality factors. We know from our HP partners that there are many theophylline tablets such as Doxofylline Tablets on the market nowadays[3]. By calculation[4], the dose of one tablet of theophylline is about 8.8 mol/L, which is much higher than the experimentally selected theophylline induction concentration. Therefore, we tentatively believe that a theophylline tablet is capable of initiating a functional switch when taken together with probiotics under a person's daily situation.

Cre-Reverse loxP system

The system consists of Theophylline-induced expression of Cre and a reverse loxP along with a promoter and RBS. To verify the viability of the system, we replaced the gene on the other side of the promoter with an mRFP. The circuit of this part is shown in the following figure 4.

To verify the success of our plasmid construction, we transformed it to DH5α and performed PCR tests, the actual electrophoresis results met the target band size. Meanwhile, we validated Cre at the protein level, as shown in Figure 2, for DH5α, with 38 kDa Cre protein bands are clearly visible in lane 7 compared to the DH5α in lane 1, as shown by the arrows.

The untransformed DH5α was used as blank control group. The target protein size was 38 kDa. The bands are the expected size, which confirmed the successful expression of our Cre protein.The vector for the validation of the Cre/loxP recombinant system has been initially constructed and the Cre expression has been validated as successful. Due to pandemic and time constraints, we are in the process of completing this part of the work and have subsequently designed an experimental protocol for the fluorescent reporter system awaiting further validation.

Degradation module

A literature review shows that the structure of tmd and dmd is very complex (see subsequent model description), so for this board we have chosen to verify the activity separately firstly. A schematic of the validation loop is shown in figure 6 below.

Trimethylamine dehydrogenase and Dimethylamine dehydrogenase

Protein gel

Firstly, we transformed the synthesized plasmids into DH5α and BL21 in sequence, where the activity of tmd and dmd was tested in a strain of BL21, the protein expression vector. We first selected a BL21 strain without any plasmid as control, and then induced the protein at 0 mmol/L, 2 mmol/L and 4 mmol/L theophylline concentrations for 9 h. The protein was then purified and run on SDS-PAGE.

Because trimethylamine dehydrogenase (TMADH) and dimethylamine dehydrogenase (DMADH) exist as dimers, their protein molecular weight would double. So, protein molecular weight of TMADH is 164.9kDa, DMADH is 166.8kDa. Based on the analysis of the experimental data, we can tentatively determine that both tmd and dmd were successfully expressed under 2 mmol/L theophylline concentration induction. There appeared to be little difference after increasing the theophylline concentration to 4 mmol/L, but the expression in the group without theophylline we suspect was a leakage of the riboswitch.

In order to increase the expression intensity of trimethylamine dehydrogenase and dimethylamine dehydrogenase, we increased the expression intensity to 4mmol/L in addition to the conventional 2mmol/L theophylline. However, as shown in the figure, there is not much difference between the induction effect of 2 and 4mmonl/L. With the help of mathematical modeling, we also understand the possible reason why the threshold of theophylline induction concentration is too low.

HPLC

Later, wetested metabolites of E.coli BL21 containing tmd or dmd through HPLC to verify that enzymes could make an effort in degrading TMA and DMA. And till now, we are still working on it.

Validation of the formaldehyde dehydrogenase gene cluster

The system consists of a formaldehyde-induced pFrmR promoter and a FrmRAB operon. In order to verify the feasibility of the system, we transformed the plasmid into DH5α and added different concentrations of formaldehyde to the bacteria for OD600 detection, with the control group set. We found that the colonies grew best under the condition of 0.001% formaldehyde, which formed a sharp contrast with the control group. At the same time, we also carried out the color reaction verification of formaldehyde and sodium sulfite fuchsin solution, and successfully saw the color change. Meanwhile, we may find the best induced concentration in the color reaction.

This is the OD600 of colonies under different formaldehyde concentrations that we measured:

This part was successfully proved.

Alleviation module

For the relief module we chose to express the short-chain fatty acid butyric acid and nattokinase, which contributes to thrombolysis. The specific circuit schematic is shown below, and the IPTG-inducible promoter was chosen to induce protein expression (Figure 11).

Butyric acid

We constructed an inducible vector pET-28a(+)-tes4 (with His-tag) by homologous recombination. The vector was first transformed into Escherichia coli DH5α for vector cloning. Then we transformed the cloning vector into Escherichia coli BL21(DE3), and then cultured it in LB liquid medium. To set up a control, we set up two groups, one group with IPTG to induce the expression of the target protein, the other group is without IPTG. Since we added a His-tag to the Tes4 gene sequence, the whole protein expressed in two groups of E.coli BL21(DE3) with pET-28a(+)-Tes4 was purified by nickel column to obtain our target protein and verified by SDS-PAGE gel electrophoresis. In addition, these two groups of bacteria were placed in a 35 ° C water bath for 48 hours, then centrifuged to take the supernatant. The supernatant was detected by gas chromatography, to measure the butyric acid. [5] Explanation of two peaks in the second picture: we suspect that Escherichia coli BL21(DE3) (with His-tag) itself will produce a substance with properties similar to butyric acid. According to the gas chromatography results of the butyric acid standard solution, it can be proved that the peak time located between 9.6min and 9.7min shows our target product, butyric acid (Figure12).

Nattokinase

The function of treatment module 2 is to produce and secrete nattokinase, dissolve thrombus forming in blood vessels, and treat atherosclerosis [6]. Since our engineered bacteria was expressed at 37℃ in intestine, we identified nattokinase and tested its fibrinolytic activity based on this.

Through SDS-PAGE analysis, we found that there was no stripe in the supernatant, which may be because we reduced the concentration of IPTG in order to produce correctly-folded nattokinase rather than inclusion bodies at 37℃, therefore the concentration of nattokinase secreted into the supernatant was low. Protein purity by His-tag showed a clear stripe at 28 kDa. At 40 kDa there was only a shallow stripe, indicating the production of correct folded nattokinase rather than inclusion bodies.

In addition, we designed and performed the assay for the fibrinolytic activity of nattokinase. According to the literature, nattokinase has fibrinolysis activity[7]. Therefore, we used a fibrin plate to test nattokinase activity. The two holes at the upper part of the plate were added 20 μL supernantant and microbial respectively, at 37 ℃ of 0.01mmol/L, as the plate shows, a transparent circle. Moreover, the control group at the lower part of plate with 20μL water, no transparent circle was found.

Suicide Module

In order to prevent engineered bacteria from affecting the environment when leaking to the environment, we designed a suicide module. Considering the room environmental temperature is lower than the human gut, we designed a temperature-sensitive suicide switch to achieve low-temperature suicide in vitro.

Referred to HZAU-China 2021 (BBa_K3733043) , we using RNA thermometer and toxin proteins whose conformation changed when exposed to 37 ℃ and has a specific RNase E binding sequence[8][9]. When transcribed into mRNA, RNase E can bind to the specific RNA sequences at 37℃ and play the role of degradation. After the toxin protein is cleaved, the bacteria can survive. At 27℃, the RBS not contained RNA could form a secondary stem-loop structure, and its follow-up sequence could be translated normally. At this point, the toxin protein behind the thermometer can be translated, acting as an RNase, and at the translational level can degrade the free mRNA in the cell, ultimately leading to cell death.

To verify the effectiveness of our suicide system, we set up two temperature gradients and designed a control group (containing only ori and Cmr) in the following experiments. After being incubated at 26℃ and 37℃ respectively for 15 hours, we could distinctly see that the liquid medium containing the toxin protein sequence was clear at 26℃, which was evidently in contrast with other conditions. At the same time, we also carried out quantitative analysis, as shown in the figure below, it was obvious that the value of the medium containing the toxin protein was lower than that of the other groups.

The safety of engineering bacteria design has always been the focus of our team. More details can be checked on "safety" page of our team.