Overview

Throughout the experiment, we used a number of measurements to verify the experimental results. First, we used fluorescence microscopy to verify the performance of the transfer switch. This allows us to observe whether the transfer switch can correctly turn on the expression of downstream gene pathways under different light/temperature conditions. In addition, in the flavor synthesis section, we used high performance liquid chromatography (HPLC) to separate and detect the presence of flavor molecules in the bacterial broth to determine whether we could synthesize the flavor substances correctly. In the suicide switch section, we measured the optical density value (OD) of the bacterial broth by enzyme standardization to monitor the growth of the engineered bacteria. In addition, to quantify the protein expression rate, we also detected the protein content in the bacterial broth using a UV spectrophotometer in combination with the BCA method.

Fluorescence microscopy

To visually verify the performance of the 2×2 transfer switch, we attached four fluorescent proteins as reporter proteins downstream of it. The engineered bacteria were cultured at 20℃/37℃ with/without blue light, respectively, and observed whether the engineered bacteria could produce the correct fluorescence by fluorescence microscopy. In the experiment, we successfully observed the expected fluorescence with a fluorescence microscope, which proved that our switch could correctly initiate the expression of the downstream gene pathway.

However, in the process of using the fluorescence microscope, we found that prolonged use of the fluorescence microscope would quench the fluorescence and affect the observation of fluorescence. Therefore, in the future design, we plan to use chromogenic proteins instead of fluorescent proteins as reporter proteins, which will avoid the disadvantages of fluorescence microscopy.

HPLC

To analyze the production of flavor substances, we used high performance liquid chromatography (HPLC) to detect the presence and concentration of vanillin and 2-phenylethanol.

In the vanillin experiment, the flow rate was 1mL/min. The mobile phase consisted of solvent A (0.1% trifluoroacetic acid in water) and solvent B (0.1% trifluoroacetic acid in acetonitrile). The following gradient elution program was used: initially, 95% solvent A+ 5% solvent B; 8min, 20% solvent A+ 80% solvent B; 10min, 80% solvent A+ 20% solvent B; 11min, 95% solvent A+ 5% solvent B. Production of L-tyrosine was monitored by measuring the absorbance at 274nm, ferulic acid at 321nm, and vanillin at 308nm. We have successfully observed the peak signals of three standard samples, but only tyrosine was detected in the culture samples(Figure 1). Further experiments need to be performed.

The production of 2-phenylethanol was determined by HPLC as well. We prepared two strategies, aiming to find the optimal elution method and thus better result. The first HPLC protocol restricted the flow rate to 1.0mL/min, and the temperature to 30 degrees Celsius. The solvent was composed of methanol and water with a constant ratio of 1:1, and we monitored the samples by testing the absorbance at 260nm. However, the result showed a deviation between the standard sample and our product in the peak signals. Next, we changed our strategy to a gradient elution program. The solvent composition changed from 80% water and 20% methanol to 50% water and 20% methanol within 5 min followed by a 5 min hold time. The flow rate was kept constant at 1.0mL/min. Under these conditions, we ultimately observed a significant peak but with a long tail, probably due to the disturbance of water-soluble impurities. Though we had confidence in the successful production of 2-phenylethanol, extraction of products and removal of impurities would bring a more reliable result.

The measures should be improved in future experiments as the curves fell short of our expectations. Several overlaps existed in the peak profiles, and the gradient elution program should be adjusted. Signals of main compositions came out just before the end of it. Moreover, preprocessing of the testing samples is essential, including the maximum possible removal of impurities to simplify the chromatography process. If possible, we might as well do some preliminary experiments to verify the existing protocols, testing the feasibility and the configuration(e.g. columns). Additionally, purifying the testing samples enables us to use the HPLC/MS method, significantly promoting accuracy and quantification.

References

Li K, Frost JW. Synthesis of vanillin from glucose. J Am Chem Soc, 1998, 120(40): 10545-10546 Ni, J., Tao, F., Du, H. et al. Mimicking a natural pathway for de novo biosynthesis: natural vanillin production from accessible carbon sources. Sci Rep 5, 13670 (2015). https://doi.org/10.1038/srep13670

Nakagawa, A., Minami, H., Kim, JS. et al. A bacterial platform for fermentative production of plant alkaloids. Nat Commun 2, 326 (2011). https://doi.org/10.1038/ncomms1327

BCA

After successfully verifying that the 2×2 transition switch can express the corresponding fluorescent protein under four different temperature/light conditions, we also quantified the expression of fluorescent protein by the BCA method to reflect the working effect of the switch more comprehensively.

Under alkaline conditions, the peptide bond in the protein can reduce Cu2+ to Cu+, and the amount of Cu2+ generated is proportional to the protein content in the solution. Each Cu2+ can form a violet-blue chelate with two molecules of Bicinchoninic Acid (BCA), which strongly absorbs light at 562 nm. The BCA method detects the concentration of protein by measuring the intensity of absorbance at 562 nm by the violet-blue chelate.

Before detecting the amount of fluorescent protein, we first need to purify the fluorescent protein. To reflect the change of fluorescent protein expression with time, we collected many samples at different time points of the bacterial broth culture. Purification of the protein by traditional affinity chromatography has many steps and is not suitable for the detection of a large number of samples. Therefore, we tried the SDS-PAGE + protein gel recovery method. After performing conventional SDS-PAGE on the samples, we stained part of the gels with Komas Brilliant Blue and used it as a reference to cut off strips of fluorescent proteins at the corresponding positions, after which the fluorescent proteins were recovered using a protein gel recovery kit. Finally, BCA working solution was added to the recovered fluorescent proteins, and the concentration was measured by detecting the absorbance at 562 nm in an enzyme marker.

Please refer to the pages of proof of concept to see the results.