Overview

Failure and progress go hand in hand. Only by learning from failure can we achieve progress. Undeniably, this applies to work in any field, especially in biological science. Working in the iGEM team, we experienced the cycles of "Design-Build-Test-Learn (DBTL)", making progress through continuous failures.

We develop two modules to produce Thaumatin steadily and efficiently in tomato fruits in our project. Our Thaumatin was optimized to remove bitterness and retain sweetness based on the directed evolution. And our production system combines a stable synthesis module with an automatic storage module, both working together to form a complete production line. Finally, we get nutritional sweetener SWEETEIN from the juice of tomato fruits straight, providing a nutritious, safe and cheap sugar substitute for the public.

In our project, we follow the principle of DBTL cycle, testing the functions of protein design, production system and sweetness analysis in the project cycle. Then to achieve the expected effect, improve and iterate based on the results.

Production of sweet protein

Expression in prokaryotic system

Design

Although Thaumatin and Brazzein are small molecule proteins^[1], the protein structure of Thaumatin contains eight disulfide bonds. We can't ensure protein will be folded correctly in the prokaryotic system, but we still made an initial attempt in E. coli. We chose the most common strain BL21(DE3) to realize fermentation production. Our designed target sequence is controlled by T7 promoter, and assembled to pET28a to get expression vector.

Build

We pick colonies from the culture medium containing kanamycin to enlarge cultivation. To prove the success of transformation, we use specific primer to amplify the target sequence from the plasmid template extracted from E. coli in PCR. Agarose gel electrophoresis proved that our expression vector successfully entered E. coli BL21 (DE3) (Fig 1).

Fig 1. The genes for Thaumatin and Brazzein were constructed on the pET28a vector

Test

Test In our experiment, the proteins expression was induced by IPTG. Then, we extracted proteins from E. coli for Western Blot verification (Fig 2). In this process, we found that the protein was very viscous when extracted, so we suspected that it was caused by the formation of inclusion bodies due to the limitations of the prokaryotic expression system of E. coli. We then broke the cells and centrifuged to get the supernatant, and found that the expression of Thaumatin and Brazzein was still very low.

Fig 2. SDS-PAGE and Western blot analyses of Thaumatin and Brazzein expression in transgenic tomato flowers. SDS-PAGE results for Thaumatin (A) and Brazzein (C), and corresponding Western blot results for Thaumatin (B) and Brazzein (D).

Learn

Yeast is one of the commonly used representatives of eukaryotic organisms in biological research and is regarded as the preferred microorganism for fermentation processes. However, through an investigation of comprehensive human practices, it has been observed that the fermentation production requires a protein purification process that is highly complex and may pose increased risks to safety. Although some studies suggest that modifying yeast can partially alter this situation, the modification of yeast strains necessitates prolonged selection and evolution to achieve stability, with the time and financial costs involved being incalculable. Clearly, yeast is not a suitable choice. Therefore, we have shifted our focus to plant-based systems, considering the practical application scenarios of sugar substitutes as food additives. We aim to save on purification steps and reduce safety risks in the process of obtaining Thaumatin. Consequently, we have chosen to conduct trials on a fruit commonly consumed in everyday life: tomatoes.

Expression in tomato

I. Transient expression

Design

Transient infection is an effective way to verify the availability of plant expression system. Many studies show transient infection can be achieved through Agrobacterium immerse or by leaf injection using a viral vector.

Agrobacterium ^[2] attaches to the surface of plant cells through its external pili. It then secretes a protein called virulence protein, which forms channels on the plant cell surface, allowing the transfer of Agrobacterium DNA into the plant cell. Considering that our chasis is tomato fruit, it is obviously not feasible to immerse the fruit in Agrobacterium to achieve instantaneous infection.

A viral infectious clone vector refers to the process where, under the action of reverse transcriptase, the RNA viral genome is transcribed in vitro to synthesize cDNA, and the cDNA is specifically modified and cloned to enable both self-replication and the expression of foreign genes. In comparison, plant viral expression vectors can remain for a longer period after leaf injection, which facilitates the detection of proteins in fruit.

Therefore, we selected the modified TRV virus for cloning. The target sequences Thaumatin and Brazzein were cloned onto the TRV2 gene of the TRV virus. The TRV vector was injected into Agrobacterium for amplification, followed by injection into tomato leaves.

Fig 3. The genes for Thaumatin and Brazzein were constructed on the pTRV2 vector

Build

We used the TRV (Tobacco rattle virus) to construct the TRV2-CaMV 35S-Thaumatin and TRV2-CaMV 35S-Brazzein plasmids and infected tomato plants. First, the plasmids were transferred into Agrobacterium GV3101 for amplification. To confirm successful transformation, we performed PCR with specific primers and verified through agarose gel electrophoresis that our expression vector had successfully entered Agrobacterium GV3101 (Fig 4,5).

Fig 4. The colonies PCR results of Thaumatin. M is the Marker. Numbers 1 to 20 are experimental groups. C is the control group. The position in the white box in the figure is the area where the target band is located.

Fig 5. The colonies PCR results of Brazzein. M is the Marker. Numbers 1 to 20 are experimental groups. C is the control group. The position in the white box in the figure is the area where the target band is located.

Test

We chose Micro-Tom as our chassis in many varieties of tomatoes. When the tomato plants reached the four-leaf stage, we injected Agrobacterium into the tomato leaves. After waiting for a period of time, we extracted RNA from tomato fruits and leaves for RT-PCR validation (Fig 6).

Fig 6. RT-PCR results of Thaumatin and Brazzein.
A: Shows the RT-PCR results of Thaumatin. M is the marker. Numbers 1 to 4 are the experimental groups. CTR is the leaf control group of wild-type tomato plants. The position in the white box in the figure is the area where the target band is located.
B: Shows the RT-PCR results of Brazzein.CTR is the leaf control group of wild-type tomato plants. Numbers 1 to 6 are the experimental groups. M is the marker. The position in the white box in the figure is the area where the target band is located.

We detected the protein expression of Thaumatin and Brazzein in positive plants. We found that Thaumatin and Brazzein can be successfully transcribed and translated in tomato plants (Fig 7).

Fig 7. Western Blot results of Thaumatin and Brazzein extracted from Micro-Tom leaves and fruits.
A: The WB result diagram of Thaumatin expression in Micro-Tom leaves. M is the Marker. Numbers 1 to 8 are the experimental groups. CTR is the WB result of wild-type tomato leaves.
B: The WB result diagram of Brazzein expression in Micro-Tom leaves. M is the Marker. Numbers 1 to 7 are experimental groups. CTR is the WB result of wild-type tomato leaves.
C: The Western Blot (WB) result diagram of Thaumatin expression in Micro-Tom fruits. M is the Marker. Numbers 1 to 5 are the experimental groups. CTR is the WB result of wild-type tomato fruits.
D: The WB result diagram of Brazzein expression in Micro-Tom fruits. M is the Marker. Numbers 1 to 5 are the experimental groups. CTR is the WB result of wild-type tomato fruits.

Learn

In the tomato expression system, Thaumatin and Brazzein were successfully synthesized, demonstrating that the plant chassis indeed has advantages for protein expression. However, from a practical application perspective, viral transient infection still has drawbacks. Even though some studies have used viral vectors for transient expression to produce various recombinant proteins, in practical production, this method still requires purification, otherwise consumers' concern cannot be alleviated. On the other hand, transient expression cannot achieve stable inheritance, making it far from suitable for long-term production. Therefore, we decided to change the method and use transgenic techniques to stably express sweet proteins.

II. GMO plant expression

Design

Transgenic technology can introduce foreign gene sequences into the host genome, causing heritable changes in biological traits. This method has long been used not only as a traditional breeding approach but also plays a crucial role in producing products through bioreactors. To ensure the sustainable production of Thaumatin and Brazzein, we decided to adopt transgenic methods. We utilized the integration ability of the Agrobacterium Ti plasmid to introduce the target gene into tomato callus tissue, achieving heterologous expression.

Build

We initially selected the eukaryotic promoter CaMV 35S for preliminary expression attempts and used the binary vector pBWA(V)HS to construct the plasmids pBWA(V)HS_Thaumatin and pBWA(V)HS_Brazzein, which were transformed into Agrobacterium GV3101. To confirm successful transformation, we performed PCR with specific primers and verified through agarose gel electrophoresis that our expression vector had successfully entered Agrobacterium GV3101 (Fig 8).

Fig 8. Colonies PCR with hygromycin gene as the target sequence. (A) Agrobacterium tumefaciens transformed with the 35S-Thaumatin-3x HA construct. (B) Agrobacterium tumefaciens transformed with the 35S-Brazzein-3x HA construct.

Test

We infected the callus tissue with engineered Agrobacterium GV3101, placed the dried explants onto co-cultivation medium, and after the differentiated seedlings grew to a certain size, cut them into sections and inoculated them onto rooting medium. The seedlings were then cultured at 23°C and finally transplanted into soil for cultivation. We collected leaf, flower and fruit samples, extracted proteins for WB analysis, and obtained positive results, indicating that transgenic tomatoes capable of stably producing sweet proteins have been successfully cultivated (Fig 9, 10, 11).

Fig 9. Western Blot results from 35S promoter transgenic tomato leaves. (A) The target band is Thaumatin-3x HA. (B) The target band is Brazzein-3x HA(B).

Fig 10. Western Blot results from 35S promoter transgenic tomato flowers. Western Blot results for (A) Thaumatin flower, (B) Brazzein flower.

Fig 11. Western Blot results from 35S promoter transgenic tomato fruits. Western Blot results for (A) Thaumatin fruit and (B) Brazzein fruit.

Learn

In this part of the experiment, we further validated the feasibility of utilizing the tomato chassis and successfully cultivated transgenic tomato plants. Therefore, we can confirm that tomatoes modified through transgenic methods can produce Thaumatin and Brazzein for use as sweeteners. However, to meet application demands, we must ensure the stability and yield of the expression process in these transgenic tomatoes. This requires us to increase expression levels and reduce degradation. As a result, we further designed a expression system and storage localization system for the next iteration.

III. Fruit-specific expression

Design

To achieve stability in the tomato production system, we considered two aspects and designed a expression system. On one hand, humans have always favored the fruit of tomato crops, with little use of leaves, stems, and other parts. This is not only because of the fruit's appealing color and high nutritional value but also due to its unique flavor and versatile applications. Therefore, we chose a fruit-specific expression promoter to reduce the metabolic burden on other parts of the plant. On the other hand, to avoid and alleviate expression inhibition caused by DNA methylation modification in tomatoes, we decided, after reviewing several studies, to use the tomato's endogenous E8 promoter^[3]. The E8 promoter is one of the most well-known fruit ripening-specific promoters and shows strong conservation across different tomato varieties. We hope that this design will enable efficient and specific expression of sweet proteins in the fruit.

Build

We inserted the Thaumatin and Brazzein genes respectively at the downstream of the E8 promoter to construct the plasmids pCAMBIA1301_Thaumatin and pCAMBIA1301_Brazzein, which were transformed into Agrobacterium GV3101. Subsequently, we performed PCR with specific primers and verified through agarose gel electrophoresis that our expression vectors had successfully entered Agrobacterium GV3101 (Fig 12).

Fig 12. Colonies PCR result(With the hygromycin resistance gene as the target band). B1-3:E8-Brazzein-1x HA. T1-3:E8-Thaumatin-1x HA.

Test

The callus tissue was infected by engineered Agrobacterium GV3101, and we used the same method and condition to cultivate transgenic tomatoes. Then, we collected samples of leaves, flowers, and fruits, extracted proteins for WB analysis, and obtained positive results. We found that Thaumatin and Brazzein were not detected in the leaves and flowers of the transgenic tomatoes but were expressed only in the fruit. This indicates that the E8 promoter can achieve the specific production of sweet proteins in tomato fruit (Fig 13).

Fig 13. Western Blot results from E8 promoter transgenic tomato leaves and flowers.
Western Blot results for Thaumatin leaf (A), Thaumatin flower (C), Thaumatin fruits (E), Brazzein leaf (B) Brazzein flower (D) and Brazzein fruits (F).

To ensure that Thaumatin synthesis induced by the E8 promoter is efficient, we further compared the expression effects of the E8 promoter and the 35S promoter. The team measured the concentration of Thaumatin in tomato fruit using a Thaumatin Enzyme-Linked Immunosorbent Assay (ELISA). A standard curve was established using Thaumatin standards, and positive plant samples for both promoters were randomly selected from WB results as the experimental group. The results showed no significant difference in expression levels between the 35S and E8 promoters, indicating that Thaumatin synthesis induced by the E8 promoter is efficient (Fig 14).

Fig 14. The expression level of Thaumatin under different promoters.

Learn

Based on the above results, we can conclude that the E8 promoter we designed has demonstrated excellent specific expression. We can ensure that Thaumatin and Brazzein are efficiently expressed only in the fruit. In the future, this production model has the potential for industrialization.

Sweet protein storage system

Design

The protein levels depend on the balance between synthesis and degradation^[4]. To increase protein accumulation, we need to ensure high synthesis and low degradation. Therefore, we must pay attention to the degradation of the protein. We chose to provide a protective area within the tomato fruit for the storage of Thaumatin, reducing the likelihood of its degradation.

In the integrated human practices, we reviewed a large number of publications and discussed our storage localization system with plant expert Prof. Mo Beixin and tomato research expert Prof. Huang Tengbo, who has years of experience in tomato studies. Based on their suggestions, we decided to use the vacuole as the storage compartment for Thaumatin, as the vacuole offers the following advantages:

• In tomato fruit cells, most of the space is occupied by vacuoles, which are often used for storing various nutrients^[5].
• The pH environment in vacuoles is lower than 7.0, and the acidic environment suits Thaumatin's acid stability characteristics.
• The environment inside the fruit vacuole is relatively more stable than the cytoplasm and apoplast during the fruit ripening process.

Therefore, we chose to use the sweet potato sporamin N-terminal propeptide (SPS-NTPP) sequence for fusion expression with the sweet protein sequence. SPS-NTPP is a kind of vacuole transform signal molecule. SPS-NTPP is a vacuolar transport signal molecule that can accurately and efficiently direct Thaumatin to the vacuole for storage. After completing its targeting function, SPS-NTPP self-degrades without causing any effects.

Build

Therefore, we chose to use the sweet potato sporamin N-terminal propeptide (SPS-NTPP) sequence for fusion expression with the sweet protein sequence. SPS-NTPP is a kind of vacuole transform signal molecule. SPS-NTPP is a vacuolar transport signal molecule that can accurately and efficiently direct Thaumatin to the vacuole for storage. After completing its targeting function, SPS-NTPP self-degrades without causing any effects.

Fig 15. Plasmid carrying the Thaumatin gene with the plant vacuole localization peptide SPS-NTTP and fluorescent protein.

Test

We transformed pGD_SPS-NTPP-Thaumatin-EGFP and pGD_Thaumatin-EGFP separately into Agrobacterium GV3101. Considering time constraints and our immediate objectives, we chose to use Nicotiana benthamiana for transient infection, and then used confocal microscopy to compare the distribution and aggregation of the fluorescence focus^[6]. The Thaumatin with EGFP guided by the vacuole localization peptide SPS-NTPP was infected into tobacco leaves by using an Agrobacterium vector. The observation results of confocal microscopy imaging showed that there were clustered green spots at the vacuole positions of tobacco leaf cells, proving that our SPS-NTPP protein can be normally expressed and function in tobacco cells (Fig 16).

In order to further verify the feasibility of the SPS-NTPP localization peptide in tomato fruits, and considering the complexity of transgenic plant cultivation, after discussion by the team, we decided to first conduct a pre-experiment of infecting leaves in tomatoes. We used Agrobacterium to infect tomato leaves with Thaumatin carrying the SPS-NTPP localization peptide and green fluorescent protein, and then continued to observe the feasibility of the localization peptide under confocal microscopy. From the figure, it can be analyzed that green fluorescent spots are clustered in the cells, and red spots are widely distributed in various positions of the cells. The green spots are caused by green fluorescent protein, while the red spots are caused by chloroplasts. Because when we designed the plasmid, we co-expressed and combined green fluorescent protein with Thaumatin, so the green fluorescent spots we see now represent the aggregation of Thaumatin. Secondly, we see that the chloroplasts of red spots are clustered around the green spots without mixing to produce yellow light, which proves that they are stored separately. Because chloroplasts exist in the cell matrix, and most of the remaining part is vacuoles, we can preliminarily infer that the heterologously expressed SPS-NTPP localization peptide in tomatoes is effective. This verifies the rationality of expressing and localizing Thaumatin in tomato fruits.

Fig 16. Confocal result diagram of verifying the vacuole localization function of SPS-NTPP in tobacco and tomatoes.

Learn

Based on the above experimental results, we can ensure that the SPS-NTPP sequence has complete functionality for targeting the vacuole. This design allows Thaumatin to be stored in the vacuole, reducing interference with Thaumatin and increasing the total yield. Therefore, we can prove that sweetein has reached the expectation of our final product, which can efficiently, stably, and specifically synthesize Thaumatin and store them in the vacuole. We can achieve the direct acquisition of Thaumatin as a nutritional sweetener in tomato juice, which demonstrates good safety and convenience in application, significantly reducing the cost of Thaumatin and providing consumers with a better sweet taste experience and convenience.