Biogenic amines can be produced by yeast through the process of fermentation. For wine consumers, these substances may cause headaches, intestinal problems, pseudo-allergic responses, and many other problems [1]. Thus, lowering the amines produced in wine production is necessary for safety concerns. Previously, efforts have been made to find better microbes and improve fermentation technic. Aiming to solve this problem more effectively, we, the safe alcohol team, started the Biogenic amine reduction Saccharomyces cerevisiae project, applying gene engineering methods innovatively.
Noticing that gene fragment AocI can conduct amine oxidase production and FAD can promote expression of the enzyme, we decided to combine them with plasmids pAO815. And we select GS115 as our vector, suitable for wine production. Our project included three basic sections-construction of recombinant plasmids, yeast conversion and protein expression, and enzyme function testing. Results showed that our products were successfully made, and the gene-engineered yeasts worked well on reducing biogenic amine production.
2.1 Amplification of fragments AocI and FAD
Fragment AocI-sized 2285 bp conducts amine oxidase production, and FAD-sized 2119 can facilitate the expression of the oxidase. We mixed corresponding primers, 2xmix, and template fragments of AocI and FAD, respectively, and put them into PCR equipment to get multiple pieces. The fluorescence fragment of the FAD product was approximately parallel to the 2000 bp marker, while the fragment of AocI was parallel to the 3000 bp one, which is plausible considering the size of the elements (Fig 1B). Thus, it proved that our PCR amplification was successful, and we prepared experiment material for the following process.
Fig 1. PCR amplification of AocI and FAD
A: PCR testing result of fragments FAD and AocI
B: Graph of FAD and AocI
2.2 Amplification and linearization of plasmid pAO815.
Plasmid pAO815 is 7709 bp in size (Fig 2B), which can function in yeast. We put a template of plasmid pAO815, 2xmix, and corresponding primers together and did a PCR to amplify and linearize it. PCR testing showed that fluorescence fragments fell approximately between 10000 and 7500 bp. Compared to the expected length of pAO815, the result convinced us that PCR amplification and linearization processes succeeded. Thus, we cut off the fluorescence fragments for the following gel purification procedure.
Fig 2 Amplification and linearization of plasmid pAO815
A: PCR testing result of pAO815 linear
B: Graph of PAO815
2.3 Construction of plasmid pAO815-AocI.
Aiming for recombinant plasmids and preparing for the following processes, we connected the two parts through homologous recombination and transformed them into DH5α. Then, applying the primers corresponding to AocI, we did a PCR testing (Fig 3A, 2285 bp). The test result showed that the size of our fragments was plausible. 100% DNA alignment (Fig 3C) further proved that our plasmid had been finely reconstructed and could successfully replicate.
Fig 3 plasmid pAO815-AocI
A: PCR testing of plasmid pAO815-AocI
B: Graph of plasmid pAO815-AocI
C: DNA alignment result
2.4 Construction of plasmid pAO815-AocI-FAD.
After we combined fragment FAD and the plasmid Pao815-AocI, we did a PCR test (Fig 4A). Compared to the markers, the size of our products aligned with the size of AocI and FAD fragments. The result showed successful conversion of the reconstructed plasmid into DH5α. DNA alignment test showed that our plasmid completely aligned with expectations within goal sequences (Fig 4C), which further proved our conclusion. *The area we requested to sequence covered the target region, so in fact the target region alignment is 100%.
Fig 4. Plasmid pAO815-AocI-FAD
A: PCR testing result of plasmid pAO815-AocI-FAD in DH5α
B: graph of plasmid pAO815-AocI-FAD
C: DNA alignment result
3.1 Yeast transformation.
We linearized plasmid pAO815-AocI-FAD and did a PCR test (Fig 5A) aligned with expectations, indicating that our plasmid reconstruction was successful.
After converting the plasmid into the yeast and conducting incubation, we tested our yeast (Fig 5C) by PCR. Fragments of FAD and AocI were respectively 2285 bp and 2119 bp. The result aligned with the size expected, showing that the plasmid had been converted into the cell and could replicate successfully.
Fig 5.
A: PCR testing result of plasmid pAO815-AocI-FAD linear
B: PCR testing result of plasmid pAO815-AocI-FAD linear in GS115
C: GS115 colonies containing plasmid pAO815-AocI-FAD in the SDS-His medium
3.2 polyacrylamide gel electrophoresis verified the size of protein bands
After product purification, we used protein electrophoresis to verify whether the major protein AocI was successfully induced in culture and checked whether AocI was successfully expressed.
As shown in Fig. 6, protein AocI has a size of 85kDa. There was a clear difference between the protein band of yeast containing the AocI plasmid and the blank control group, indicating that the AocI gene was successfully expressed.
At the same time, the protein banding was most clearly observed at 0.50% methanol. It can be seen that the optimal environment for AocI expression was 0.50% methanol. (Fig 6)
Fig 6. Result of polyacrylamide gel electrophoresis
3.3 Validation map of amine oxidase functional activity
The graph shows changes in crude and purified AocI protein concentrations at different methanol-induced concentrations. It can be intuitively seen from the table that the crude protein concentration was the highest at 0.50% methanol-induced concentration. The lowest crude protein concentration was observed at the induced concentration of 1.00% methanol. After 0.50% methanol, crude protein concentration decreased with increasing methanol-induced engagement. There was little difference in protein concentration after purification.
Fig 7. Validation map of amine oxidase functional activity
4.1 Using the acid-base indicator method to qualitatively compare the ability of the transformants to metabolize biogenic amines
Bromocresol purple, as an acid-base indicator, turns yellow in acidic conditions and purple-red in alkaline conditions. Therefore, after adding histamine to the system, the culture medium will exhibit a purple-red color. When histamine is hydrolyzed by oxidases, the products are ammonium salts and aldehydes, which are weakly acidic. As a result, the alkalinity in the hydrolysis area decreases, gradually changing to yellow. Thus, it is possible to determine whether histamine has been oxidatively hydrolyzed by observing the color change.
In order to evaluate the activity of amine oxidase expressed by our yeast, bromocresol purple and different histamine concentrations were added to the PDA solid medium. For comparison, transformants and blank control (wild strains) were inoculated on the culture medium.
After a certain period, the color of the bromocresol purple in the plates changed from purple to yellow, and the color change of plates within the transformants was more prominent. At the same time, the diameter of the transparent circle around the transformants was longer than that of the blank control group. It can be seen that the transformants have a more vital ability to metabolize biogenic amines.
Fig 8. results of the transparent circle experiment and the control group.
4.2 Determination of histamine content using High-Performance Liquid Chromatography (HPLC)
High-performance liquid chromatography, or HPLC (High-performance liquid chromatography) is a widely used separation and detection technique in analytical and preparative chemistry. It uses a mobile phase to separate and detect chemical samples from a stationary phase.
To further investigate the ability of the strain to degrade histamine, we also performed an HPLC assay (Figure 9). Histamine concentrations were measured at five different temperatures (20, 25, 30, 37, 45℃) and different sampling times (12, 24, 36, 48, 72 h).
We first incubated the bacterial solution to OD 0.6-0.8 to give the bacteria the highest viability. Then histamine was added and the initial concentration of histamine was 200ug/ml. changes in histamine content were detected using HPLC. As can be seen from the chromatogram in Fig 9, compared with the control group GS115, pAO815-AocI, and pAO815-AocI-FAD, the measured time of histamine peak was around 18-19 min. According to the calculations, it was learned that the concentration of histamine was significantly decreased after 72h of continuous cultivation, and the pAO815-AocI-FAD strain had a more significant decrease in histamine concentration than the pAO815-AocI strain, which indicated that the pAO815-AocI-FAD strain had a better ability to degrade histamine, and this is in line with the result of the previous clear circle experiments.
Fig 9. Results of the HPLC.
A: Chromatograms of GS115, pAO815-AocI and pAO815-AocI-FAD at 12h
B: Chromatograms of GS115, pAO815-AocI and pAO815-AocI-FAD at 72h
According to the image results, the content of histamine decreased gradually with time, and the fastest degradation of histamine was observed at 30℃. The lowest content of histamine was observed at 72h. A pattern can be observed: all show monotonically decreasing, but obviously, the rate of decrease at 30℃ is accelerated, and also after 73h of experiment, the lowest values.
Fig 10. Comparison of histamine content values at different temperatures.
5. Next Experiment Plan
The yeast we used in this experiment is not the common Saccharomyces cerevisiae, so we will use the more common Saccharomyces cerevisiae for further investigations. With our instructor’s assistance, we drew a preliminary experiment plan to construct the recombinant Saccharomyces cerevisiae.
5.1 Plasmid Construction
We chose pYES as the vector because this Saccharomyces cerevisiae is more commonly used. Then, the three essential parts of the experiments, namely recombinant plasmid construction, yeast transformation and protein expression, and enzyme function detection, were carried out again.
Plasmid pYES is 5857 bp in size (Fig 9), which can function in yeast. We put a template of plasmid pYES, 2xmix, and corresponding primers together and did a PCR to amplify and linearize it.
Fig 11. Graph of pYES2
Aiming for recombinant plasmids and preparing for the following processes, we connected the two parts through homologous recombination and transformed them into AQ. Then, applying the primers corresponding to AocI, we did a PCR testing (Fig 10).
Fig 12. Graph of plasmid pYES-AocI
After we combined fragment FAD and the plasmid pYES-AocI, we did a PCR test (Fig 11). Compared to the markers, the size of our products aligned with the size of AocI and FAD fragments.
Fig 13. Graph of plasmid pAO815-AocI-FAD
5.2 Amine Oxidase Activity Test
In the end, we will validate amine oxidase functional activity again since we change the host.
5.3 Fermentation Test
In order to simulate the real application context, we will conduct a mini fermentation experiment with our recombinant Saccharomyces cerevisiae to produce alcohol. In this part, we will evaluate the alcohol production ability of our recombinant Saccharomyces cerevisiae for commercial consideration and the amine level during the fermentation compared to the wild.
5.4 Future Plan
Biogenic amine-reducing Saccharomyces cerevisiae will be extended to other fermented foods with high biogenic amines, such as canned fish products, fermented sausage, and cheese, to broaden its application scope further. Besides, we will also dig into investigating amine oxidase's biological activity and biogenic amine's degradation mechanism to provide a theoretical basis for applying amine oxidase[1].
[1] Niu Tian Jiao. THE MICROBIAL FLORA STRUCTURE AND DEGRADATION OF BIOGENIC AMINES DURING RICE WINE BREWING [D]Harbin Institute of Technology,2020