2×2 transfer switch

Validation by fluorescent proteins

To verify the feasibility of the switch, The UCAS-China iGEM team 2022 designed two plasmids (pET-28a) containing our switch sequences. With four different fluorescent proteins added as reporters, we can quantify the feasibility of our design.

We designed two plasmids (pET-28a) containing our switch sequences to verify their practicability. After managing to transform our plasmids into the competent cell (DH5α), We cultured them in four circumstances respectively. To maintain the temperature, we placed the conical flasks(500mL) filled with the bacterial solution(200mL) in a temperature constant shaker, and control the temperature at 25℃ and 37℃. To guarantee the light variation, we used tinfoil to cover one of the conical flasks, and used blue-colored transparent wrapping up the white illuminant and the euphotic form of the shaker. After 16,20,36h cultivating, we took out 15μL bacterial fluid to make a temporary slide and observed it through the fluorescence microscope in the stimulating wave listed below.

In the absence of blue light, YFP (yellow) and EYFP (cyan) proteins will be produced at 37°C and 25°C respectively. In contrast upon blue light excitation, the reporter located at the other side of the switch will separately fabricate mCherry (red) and JGFP (green) proteins at 37°C and 25°C. We transformed our plasmids into the competent cell (DH5) and successfully observed the expression of the report proteins (Results are as follows). To control for fluorescence, untransformed DH5 incubated in LB media was used as a blank control.

BCA protein concentration measurement method:

After verifying that the 2×2 transition switch can express the corresponding fluorescent proteins under different light/temperature conditions, we also measured the expression of four fluorescent proteins by the BCA method to further demonstrate the function of the 2×2 transition switch.

To reflect the change of fluorescent protein expression over time, we needed to collect many samples at different time points. Therefore, instead of using the traditional affinity chromatography method to purify the fluorescent proteins, we chose the SDS-PAGE + protein gel recovery method to obtain the fluorescent proteins. After SDS-PAGE of the total protein samples of the bacterial broth obtained at different time points, the bands corresponding to the positions of the fluorescent proteins were cut off and the fluorescent proteins in the SDS-PAGE were extracted using a protein gel recovery kit. Finally, the concentration of the recovered proteins was quantified by BCA method.

In the process of protein concentration determination by BCA method, we configured different concentrations of samples with standard protein samples, treated with the BCA kit and placed under the enzyme standard to detect the absorption intensity at 562 nm, so as to plot the standard variation curve of protein concentration versus absorbance (as shown in the figure below).

After that, the samples were processed and the absorption intensity was detected at 562 nm. The results were as follows.

Flavor Substance

Vanillin

The pathway to vanillin is complicated, and analyzing with High Performance Liquid Chromatography(HPLC) is a new challenge for us. Thus we have partially validated the function of the first plasmid(BBa_K4256011). Further experiments are still needed to verify the functions of the other two plasmids. 

In Figure 1 are time-intensity curves of HPLC analysis from the background(ddH2O), a standard sample of L-tyrosine, and culture samples of strains with and without the plasmid(BBa_K4256011). Signal intensity represents the absorbance at 274nm. We can conclude that the retention time of L-tyrosine is 2.5min by comparing two curves of the background and the standard sample. The peak signal intensity from samples of strains with plasmids is remarkably higher than that from the control group at around 2.5min, indicating that L-tyrosine is successfully overproduced.

2-Phenylethanol

The experiment was a preliminary success, but we have to admit that this conclusion is not convincing enough for there was an excess of interfering substances in the testing process. Selective test methods for tyrosine are required for future studies. Also, the culture samples were collected 21 hours after the addition of IPTG, which is too long for cooking a meal. The production of L-tyrosine in a limited time frame should be evaluated in further investigation.

Our plasmids carry His tag at both the N and C termini of the polycistron, so we purified the target proteins by Ni affinity chromatography and examined the purification results using SDS-PAGE. The protein purification and detection results of the broken supernatant of the 2-phenylethanol-synthesizing strain are shown in Figure 2. At the concentration of 50 mM and 100 mM imidazole, there were clear bands and a very small amount of foreign protein. The bcaT (about 50-60 kDa) is located between 55 and 70 kDa. The narrow band on the top is ADH. This result can indicate that our protein was successfully expressed and the multiple cis-trans expression was normal - there is a difference in the expression of the two proteins.

Moreover, 2-Phenylethanol could be synthesized from phenylalanine, and 2-phenylethanol could be detected through its special structure of benzene ring, which possess characteristic peak at the wavelength of about 254 nm. HPLC/MS results demonstrated that there exists strong peaks at the wavelength of 254nm.(Figure 3) However, the peaking time remained to be different. After confirming the sample itself did emit a strong smell of rose petals - which indicated the true existence of 2-phenylethanol, we assumed that the concentration may be too little, and it mixed with the substrates, and other impurities, which might also occupy the structure of benzene ring, leading it to have similar characteristic peak at the wavelength of 254nm. However, it still supported the existence of 2-phenylethanol, comparing with the standard samples (5% and 10%)。

3-methylbutanal

Since the plasmid carries His-tag at both the N and C termini of the multi cis-transons, purifying the target proteins by Ni-affinity chromatography and examining the purification results using SDS-PAGE are effective and persuasive. The results of protein purification and examination of the fragmented supernatant of the strain synthesizing 3-methylbutyraldehyde are shown in Figure 4. Theoretically, bcaT and gdh expressed proteins with molecular weights of about 60kDa and 178kDa. At the concentration of A fluid(5mM imidazole), there was a clear band at about 60 kDa. This result can indicate that our protein is successfully expressed and the expression of polycistron is verified, but there do exist differences in the expression of three proteins, which leads to different width on SDS-PAGE.

Functional substance

Expression of Amuc_1100 and ovalbumin

When designing the plasmids, we added 6xHis tags to the corresponding positions of the C-terminus of the Amuc_1100 protein and ovalbumin protein sequences in the engineered bacteria and tried to induce the expression of Amuc_1100 and ovalbumin with IPTG after culturing the engineered bacteria. After lysing the engineered bacteria, we subjected the lysate supernatant to nickel column affinity chromatography, and the products were analyzed by SDS-PAGE to obtain the corresponding bands of Amuc_1100 protein and ovalbumin protein, which proved that these two functional substances could be successfully expressed.

Expression of Rubisco

When designing the plasmid, we added a 6xHis tag at the position corresponding to the C-terminus of the Rubisco protein sequence in the engineered bacteria and tried to induce its expression with IPTG after culturing the engineered bacteria. After lysing the engineered bacteria, we subjected the supernatant of the lysate to nickel column affinity chromatography, but we did not obtain a band corresponding to Rubisco protein after SDS-PAGE analysis of the product. We also took the total lysate protein and the lysate precipitate for SDS-PAGE analysis, and both obtained bands corresponding to Rubisco protein. We speculated that this was because Rubisco protein was of eukaryotic origin and it would form inclusion bodies into the lysate precipitate when it was expressed too fast and too much under the control of T7 promoter in the engineered bacteria. In conclusion, our experiments demonstrated that Rubisco protein could be successfully expressed in our engineered bacteria. We expected that the expression rate of Rubisco protein would decrease in the future when placed under the regulation of the 2×2 transfer switch, and inclusion bodies might no longer appear.

Suicide switch

Validation of ccdB toxicity

We incubated the E. coli strain with the ccdB-pET-28a(+) plasmid until mid-log growth, then added IPTG solution to induce ccdB protein expression ( the concentration of IPTG in the bacterial solution was 1 mg/mL). Another group of the same E. coli (mid-log growth) with the ccdB-pET-28a(+) plasmid was set as a control, to which deionized water equal to the IPTG solution was added. Starting from the addition of IPTG (0min), we measured the OD value at 600nm (OD values are positively correlated with the bacterial concentration in the solution) for both groups at regular intervals, and the toxic effect of ccdB was detected by comparing OD value in the two groups.

The data are shown in Table 2, and the optical density change curve is in Figure 20. We can see that the bacterial concentration of the group with IPTG-induced ccdB expression decreased significantly after a period of increase; the initial bacterial concentration of the group with deionized water was substantially lower than that of the group with IPTG, but after 30 minutes of incubation, the bacterial concentration began to exceed that of the group with IPTG, and after 240 minutes it far exceeded that of the group with IPTG. Considering that protein expression takes some time, our experimental results can fully prove the practical killing effect of the ccdB toxin protein.

Validation of Oxygen-controlled Suicide Switch (OCS)

Using the oxygen control system FNR-HIP-1 and the TA system CcdAB as the basis for our oxygen control switch was a challenge. Our validation was halfway through for the oxygen-controlled promoter HIP-1 under the impact of the epidemic due to experimental conditions and time constraints: it was not available to respond under normoxic conditions to initiate expression of downstream genes. This experiment is necessary to ensure that the engineered bacteria cannot survive in a normal environment. We are still missing a control experiment: it can respond under anaerobic conditions to initiate downstream gene expression. This control experiment is essential to ensure that in the future our chassis bacterium Lactobacillus delbrueckii subsp. can survive and express the product properly under anaerobic conditions.

We used pET-28a(+) as the vector and first used E. coli TOP10 as the chassis bacterium for initial validation. As shown in Figure 21, we can only show for the time being that it does not respond to initiate the expression of downstream genes under normoxic conditions (no red bacterial colony).

Theoretically, the expression of the OC-eforRed element is sufficient for TOP10 to produce the red fluorescent protein eforRed visible to the naked eye as long as fully anaerobic conditions were achieved. But due to the epidemic and time, we only did an anaerobic experiment(the oxygen concentration was around 0.5%) in an anaerobic incubator and anaerobic tank but didn't successfully verify that it could respond to initiate the expression of downstream genes under anaerobic conditions.

We will then focus on controlled experiments to initiate downstream gene expression in response to anaerobic conditions of the oxygen-controlled promoter HIP-1, and on experiments to validate the feasibility of the complete gene sequence of the suicide switch. We then intended to modify the complete sequence gene circuit to allow separate isolation and purification of the FNR, CcdA, and CcdB proteins produced during the process so that the proteins could be identified. Also retaining the lactose manipulator in the laboratory sequence allowed the toxin gene ccdB to be induced by IPTG, but removing the lactose manipulator in the actual application sequence allowed it to be used directly for oxygen-controlled suicide.

The greatest difficulty we need to solve is the implementation of anaerobic conditions. We would like to solemnly thank Ms. Jin Zhong's lab at the Institute of Microbiology, Chinese Academy of Sciences.

Growth curve determination of Lactobacillus delbrueckii subsp.

To verify that our selected final chassis bacterium Lactobacillus delbrueckii subsp. could grow normally under anaerobic conditions to express the product for meal replacement products and that its growth is inhibited under normoxic conditions to further reduce its escape risk, we performed 3h and 8h growth curves in a liquid medium under normoxic and hypoxic culture conditions, respectively The growth curves were measured in liquid medium under normoxic and hypoxic conditions. We used an enzyme standard to measure OD600 to reflect the bacterial concentration. The experiments were recorded in Table 3 and Table 4, and the curves were plotted in Figure 23 and Figure 24.