Team:NAU-CHINA/Engineering
ENGINEERING
LOOP 1
Design
Selenium is one of the essential elements for the human body. Although trace of selenium is necessary for cellular function in many animals, including humans, both elemental selenium and (especially) selenium salts are toxic in even small doses, causing selenosis.
Seriously uneven spatial distribution is one of the most prominent characteristics of selenium. In the vicinity of mines and factories, the selenium content especially in the form of selenate and selenite, is significantly higher than in other areas, causing serious damage to animals living nearby. The severe situation is mainly due to the fact that selenium is dispersed into the environment along with the mining and processing of various metal minerals, but no one has taken it seriously.
Although natural selenium cycle system could relief selenium pollution to some extent, there still exists conspicuous imbalance of selenite distribution in China. The main cause of selenium pollution is selenate (SeO42-) and selenite(SeO32-). Therefore, NAU-CHINA is committed to solving selenium pollution in the environment, by transforming selenite and selenate into selenium nanoparticles that can be used in many fields, such as foliar fertilizer, antibacterial and so on.
Fig.1 Global selenium cycle in nature
By investigating literatures, we found that many microorganisms have the ability to reduce selenite and produce SeNPs, such as Providencia rettgeri HF16-A and our chosen bacteria E.coli. Although it is consistent with our original intention to solve selenium pollution, the mechanisms of reduction processes in these strains are not clear. For example, in HF 16-A, GSH and the bacteria's own partial reductase are involved in the process of selenite reduction, but these strains are not commonly accessible in the laboratory. Plus, considering biosafety issues, we didn’t choose these microorganisms as chassis. Our team hopes to use synthetic biology methods to engineer the process of reducing selenite, which can be applied to most strains. We tried to improve the ability of engineered E.coli to produce SeNPs. Eventually, we found that there are two main processes in the treatment of selenium pollution.
Ⅰ : Selenite and selenate is reduced to Se0.
Ⅱ : Se0 is wrapped by intracellular substances to form selenium nanoparticles.
As mentioned above, most microorganisms can reduce selenate and selenite, so we set our sights on the second step(Ⅱ)--the package of Se0. We found that the SefA protein from Thaurea selenatis (a β-proteobacteria) can efficiently and stably wrap the Se0 atoms into SeNPs with a size of about 100-300nm. Also, literature had pointed out that by expressing SefA protein, the tolerance of bacteria to selenate and selenite will increase. So we constructed sefA gene(BBa_K3735001) as our first part.
Next, in the selection of proper chassis, we noticed E.coli MG1655 with a strong biomineralization ability, and the strain had been used for selenium nanoparticles production experiments. However, due to the epidemic and biosecurity restrictions, we were unable to obtain MG1655 strains. So, we made a substitution, hoping to verify our parts in E.coli BL21 first.
Furthermore, as the metabolic burdens of chassis could be influenced by protein expression of different intensities, to express sefA, we designed three promoters of different intensities to find the most suitable promoter intensity. Furthermore, considering in the real world, the content of selenite in different environments is very different, so we conducted a set of concentration gradient experiments to find the optimal combination of protein expression intensity and selenite concentration.
| Concentration(mM) | 0 | 3 | 6 | 9 | 12 | 15 | 18 | 21 | 24 |
|---|---|---|---|---|---|---|---|---|---|
| Promoter Strength | Low(BBa_J23105) | ||||||||
| Middle(BBa_J23101) | |||||||||
| High(BBa_J23119) | |||||||||
Table 1 Concentration gradient experiment (Use sodium selenite to prepare different concentrations, and use pSB1C3 plasmid backbone to express the elements we designed)
Build
In the construction of the sefA expression system, we followed the RFC10 rules and constructed modular parts to facilitate the construction of the final expression devices. We obtained the sefA gene and the promoters through complete synthesis and chose E.coli BL21 as the chassis organism which will not cause biological hazards technically. If you want to know the detailed experiment protocols, you can visit our protocolpage.
Test
First, we tested parts we constructed through sequencing verification. The results showed that we have successfully constructed the following components: J23105-sefA (BBa_K3735003), J23101-sefA (BBa_K3735004), J23119-sefA (BBa_K3735005).
Fig.2 Plasmids(J23105-sefA (BBa_K3735003), J23101-sefA (BBa_K3735004), J23119-sefA (BBa_K3735005).)
Next, we conducted a set of concentration gradient experiments on the basis of getting the optimal components. The results have been shown in the Fig.3.
Fig.3 Concentration gradient phenomenon diagram
Learn
This set of experiments showed that our project was very promising. However, we didn’t found any indicator to reflect the output rates mathematically, only through the combination of different concentrations and different strength promoters, we could roughly distinguish the groups with the highest yield from the color. We found that when the concentration of selenite was 12 mM, 15 mM, and 18 mM, the test tubes of E.coli with three different expression intensities had the darkest red color, which implied that the yield of SeNPs was the highest at these three concentrations. We need further experiments to explore the most suitable combination of concentration and intensity and find a standard to measure the yield of SeNPs.
Later, we found that the absorption peak of SeNPs detected in most articles was around 300nm. We combined multiple literatures and selected them according to the actual situation. OD370 was temporarily used as an index to measure the final output. It is worth noting that this only served as a temporary indicator during our experiment, and we were still looking for a more accurate method to measure the output.
LOOP 2
Design & Build
According to the pre-experimental results, we selected the above three different concentrations of selenite (12, 15, 18mM) for more detailed characterization experiments. According to the method in the literature, we added selenite of planned concentrations to three different types of Escherichia coli cultured in a 50 mL Erlenmeyer flask. Each flasks were sampled in a 96-well plate every two hours. Afterwards, OD600 and OD370 were tested to obtain the growth of bacteria and the possible output of SeNPs.
Test
After the bacterial culture, 9 samples were detected for OD600 and OD370. The results were demonstrated below.
Fig.4 Growth curves of bacteria expressing different intensities of sefA at different selenite concentrations. (Blue: 12mM Na2SeO3; Orange: 15mM Na2SeO3; Green: 18mM Na2SeO3; Square(█): Low intensity; Triangle(▲): Middle intensity; Inverted triangle(▼): High intensity )
Fig.5 The curve of the absorbance of different samples at 370nm. (Blue: 12mM Na2SeO3; Orange: 15mM Na2SeO3; Green: 18mM Na2SeO3; Square(█): Low intensity; Triangle(▲): Middle intensity; Inverted triangle(▼): High intensity )
Learn
After the above experiment, we found that the expression of SefA protein can significantly improve E.coli tolerance to selenite. It is possible that SefA increases the efficiency of encapsulating selenium into SeNPs, and strengthens the reduction ability of E.coli. However, our engineered bacteria still has a certain gap in the reduction rate compared with specific strains screened from selenium-enriched soil. Therefore, we tried to find a way to improve the reduction ability of E.coli to reduce selenite. We reviewed the literature on the reduction pathways of selenate and selenite. The mechanism of reduction was listed as follows.
When selenate in the environment enters the periplasm, it is reduced to selenite through the reactions of a series of enzymes (SerABC), and then, selenite enters the cell and is reduced to elemental selenium. The reduction of selenite is the most important procedure in the whole reaction chains. Therefore, we focused on the improvement of the reduction rate of the engineered bacteria to deal with selenium pollution.
Fig.6 Model showing the proposed electron transport pathway involved in SeO42- and SeO32- reduction
In the study of Xian Xia et al., they discovered that NAD(P)H dependent FMN reductase-CsrF (Chromate selenite reductase flavoenzyme) from Alishewanella sp. WH16-1, which can reduce selenite to elemental selenium. In the follow-up process of searching for this gene, we discovered that E.coli BL21 itself has an FMN reductase with similar function—SsuE, which indicates that this enzyme may be involved in the reduction of selenite. So we hope to increase the final output by increasing the expression of ssuE.
In the meantime, when exchanging ideas with Professor Wenming Zhang from Nanjing Tech University, Professor Zhang advised that we could directly use E.coli BL21 for the follow-up experiments. Professor Zhang pointed out that E.coli BL21 and MG1655 may not have much difference in reduction efficiency. After our in-depth thinking on the subject, we changed from MG1655 to BL21 for all subsequent experiments.
LOOP 3
Design
In the construction of SsuE module, we still designed three parts with different strengths. By testing the growth status of strains with different ssuE intensities and different concentrations of selenite, we tried to strike a balance between reduction efficiency and bacterial metabolic burden.
Before we construct the ssuE expression part, considering that SsuE is a kind of NAD(P)H dependent FMN reductase, which is involved in the intracellular metabolic process and related redox reactions, we reviewed the literatures concerning similar heterologous expression. We found that bacteria go through severe metabolic stress during the expression of these proteins. Meanwhile, there is always a limit to the expression of these enzymes. Therefore, we also use mathematical model to simulate the protein expression. To explore the problems we might encounter in the process of part construction. Through the analysis of the growth status in the model, we might obtain valid information to select the appropriate promoter intensity to express SsuE protein. As the famous idea illustrated, “Model directs experiment, experiment reflects model”.(more details see SP model)
Through mathematical simulation of the ssuE expression, we found that due to the metabolic pressure caused by the reductase on the bacteria, ssuE expressed with high intensity and medium intensity severely inhibited the growth of the bacteria after being incubated for hours.
Build
We obtained the ssuE gene by colony PCR and combined it with three promoters with different strength to build the SsuE expression device. At the same time, we still employed the pSB1C3 plasmid as backbone.
Test
In the detection of expression elements, we found that only low expression devices met our expected results while the promoters of other devices had different levels of mutations. This result is also in accordance with the model’s metabolic pressure prediction. Here, we had verified the model through experiments.
Both mathematical model and experimental analysis proved that only the low-level SsuE could be successfully realized. Therefore, we employed this device to carry out OD600 and OD 370 detection and get the following results. (more details see Results>Test of Bacterial Density and SeNPs Production)
Learn
Facing the above experimental results, we tried to search for certain explanations from the literature. It is believed that as a reductase, the burgeon of SsuE would break the bacteria's own redox balance and disrupt the electron transport chain. To resist this change, bacteria initiated a self-protection mechanism and reduced the expression of ssuE. But fortunately, our characterization of SsuE still worked. Under OD370 conditions, we clearly observed that the low-level of ssuE could also improve the reduction efficiency of selenite.
LOOP 4
Design
After successfully constructing three different intensities of sefA expression devices and ssuE low-level device, we combined the two devices in bacteria for co-expression. By constructing a co-expression system of ssuE and sefA, we could not only improve the stability of SeNPs, but also enhance selenite reduction in the engineered bacteria. Therefore, we designed a combination of three sefA with different intensities and low-level ssuE expression device.
Considering that the particle diameter of SeNPs produced by bacteria could make a great difference to its bioactivity and antibacterial properties. Therefore, we tested differences between the SeNPs produced in the three strains in particle size. We planned to lyse and lyophilized the bacteria solution after incubation for 12 hours to extract SeNPs, and finally measured its particle diameter in scanning electron microscopy (SEM).
Build & Test
In order to construct the ssuE-sefA expression system, we constructed standardized sefA and ssuE devices on the basis of the existing materials, double-digested the two standardized devices, and finally achieved DNA ligation with T4 DNA ligase.
After construction of the 3 co-express systems with different intensities, we utilized SEM to characterize the SeNPs, and the results can be seen Results>Electron microscopic characterization of SeNPs.
Fig.7 SEM results of SeNPs production by different intensity sefA.(A: SeNPs produced by BBa_K3735888; B: SeNPs produced by BBa_K3735014; C: SeNPs produced by BBa_K3735011)
As shown in the figure above, we found that after the co-expression of ssuE and sefA, the SeNPs produced by strains of the high and middle level have a larger particle size compared with the low-level expression strains in terms of particle diameter. According to what had been found in literatures, the larger the particle size of SeNPs, the worse their quality. Therefore, we finally selected low ssuE and low sefA co-expression strains for the subsequent optimization experiments.
Design
After the selection of expression system, we started to search for the most suitable conditions for our engineered bacteria incubation and maximum output efficiency. Therefore, we took advantage of the orthogonal experiment scheme designed by our Model and established the corresponding response surface to optimize the three aspects of pH, temperature, and concentration. Again, we corresponded with the idea: “Model directs experiment, experiment reflects model”.
Regarding the measurement of SeNPs production, later on we found that there still existed conspicuous error in measuring the concentration of SeNPs by detecting the absorbance under OD370. This is mainly because the absorption peaks of SeNPs are quite different when obtained by different methods. And for the existence of various substances in SeNPs whose relative molecular mass is difficult to determine, we aimed to set a more reasonable measurement unit to measure the output of SeNPs. Fortunately, inspired by the "input-output ratio" in economics, we proposed g(SeNPs)/g(Na2SeO3) measurement unit, which represented the proportion of the input selenite and the output SeNPs, to measure the production efficiency of SeNPs under different conditions. This unit could offset complicated situations like different concentrations of selenite medium containing different amounts of selenite, and the production of different amounts of SeNPs under different culture conditions.
| Number | Temp（℃） | Concentration（mM） | pH |
|---|---|---|---|
| 1 | 30℃ | 12mM | pH=6 |
| 2 | 16℃ | 12mM | pH=7 |
| 3 | 37℃ | 12mM | pH=7 |
| 4 | 30℃ | 12mM | pH=8 |
| 5 | 16℃ | 15mM | pH=6 |
| 6 | 37℃ | 15mM | pH=6 |
| 7 | 30℃ | 15mM | pH=7 |
| 8 | 30℃ | 15mM | pH=7 |
| 9 | 30℃ | 15mM | pH=7 |
| 10 | 30℃ | 15mM | pH=7 |
| 11 | 30℃ | 15mM | pH=7 |
| 12 | 37℃ | 15mM | pH=8 |
| 13 | 16℃ | 15mM | pH=8 |
| 14 | 30℃ | 18mM | pH=6 |
| 15 | 16℃ | 18mM | pH=7 |
| 16 | 37℃ | 18mM | pH=7 |
| 17 | 30℃ | 18mM | pH=8 |
Build
The building process was similar to the Build step in Loop 4. During the construction of the ssuE-sefA co-expression system, we strictly complied with the RFC10 rules.
Test
First, we tested the constructed devices through sequence verification and successfully built the pSB1C3-J23105-ssuE-J23105-sefA.
At the same time, we examined different samples from the response surface experiment (More details see CO Model), and the phenomenon of figure is shown below(Fig.8).
Fig.8 Response surface experiment of SeNPs synthesis.
Learn
We used response surface model to simulate the optimal SeNPs production conditions (details see Model). Finally, we obtained the optimal conditions and maximum yield as follows：
| Temperature | Selenite Concentration | pH | SeNPs Yield |
|---|---|---|---|
| 30.26 | 14.33 | 6.75 | 0.0315562 |
By response surface model analysis, we simulated the optimal experimental conditions of SeNPs production: temperature at 30.26°C, selenite concentration at 14.33mM, pH at 6.75. Finally, the optimal SeNPs yield of 0.0315562g was obtained.
Nancharaiah Y V, Lens P N. Ecology and biotechnology of selenium-respiring bacteria[J]. Microbiol Mol Biol Rev, 2015, 79(1):61-80.
Santos S, Ungureanu G, Boaventura R, et al. Selenium contaminated waters: An overview of analytical methods, treatment options and recent advances in sorption methods[J]. Sci Total Environ, 2015, 521-522:246-260.
Lemly A D. Aquatic selenium pollution is a global environmental safety issue[J]. Ecotoxicology and Environmental Safety, 2004, 59(1):44-56.
Nancharaiah Y V, Lens P N. Selenium biomineralization for biotechnological applications[J]. Trends Biotechnol, 2015, 33(6):323-330.
Ikram M, Raja N I, Javed B, et al. Foliar applications of bio-fabricated selenium nanoparticles to improve the growth of wheat plants under drought stress[J]. Green Processing and Synthesis, 2020, 9(1):706-714.
Bartůněk V, Vokatá B, Kolářová K, et al. Preparation of amorphous nano-selenium-PEG composite network with selective antimicrobial activity[J]. Materials Letters, 2019, 238:51-53.
Omran B A. Prokaryotic Microbial Synthesis of Nanomaterials (The World of Unseen) [M]. Springer International Publishing. 2020: 37-79.
Avendano R, Chaves N, Fuentes P, et al. Production of selenium nanoparticles in Pseudomonas putida KT2440[J]. Sci Rep, 2016, 6:37155-37155.
Fernandez-Llamosas H, Castro L, Blazquez M L, et al. Speeding up bioproduction of selenium nanoparticles by using Vibrio natriegens as microbial factory[J]. Sci Rep, 2017, 7(1):16046-16046.
Huang S, Wang Y, Tang C, et al. Speeding up selenite bioremediation using the highly selenite-tolerant strain Providencia rettgeri HF16-A novel mechanism of selenite reduction based on proteomic analysis[J]. J Hazard Mater, 2021, 406:124690-124690.
Bulgarini A, Lampis S, Turner R J, et al. Biomolecular composition of capping layer and stability of biogenic selenium nanoparticles synthesized by five bacterial species[J]. Microb Biotechnol, 2021, 14(1):198-212.
Butler C S, Debieux C M, Dridge E J, et al. Biomineralization of selenium by the selenate-respiring bacterium Thauera selenatis [J]. Biochem Soc Trans, 2012, 40(6):1239-1243.
武童, 汪振炯, 华春, et al. 基于大肠杆菌MG1655载体矿化合成纳米硒的研究[J]. 南京工业大学学报(自然科学版), 2017, 39(03):52-57.
Mollania N, Tayebee R, Narenji-Sani F. An environmentally benign method for the biosynthesis of stable selenium nanoparticles[J]. Research on Chemical Intermediates, 2015, 42(5):4253-4271.
Al-Kurdy M J. Synthesis and Characterization of Black Currant Selenium Nanoparticles (Part I)[J]. The Iraqi Journal of Veterinary Medicine, 2020, 44(2):25-34.
Cittrarasu V, Kaliannan D, Dharman K, et al. Green synthesis of selenium nanoparticles mediated from Ceropegia bulbosa Roxb extract and its cytotoxicity, antimicrobial, mosquitocidal and photocatalytic activities[J]. Scientific Reports, 2021, 11(1):1032-1032.
Dhanraj G, Rajeshkumar S. Anticariogenic Effect of Selenium Nanoparticles Synthesized Using Brassica oleracea [J]. Journal of Nanomaterials, 2021, 2021:1-9.
Xia X, Wu S, Li N, et al. Novel bacterial selenite reductase CsrF responsible for Se(IV) and Cr(VI) reduction that produces nanoparticles in Alishewanella sp. WH16-1[J]. J Hazard Mater, 2018, 342:499-509.
Driggers C M, Dayal P V, Ellis H R, et al. Crystal structure of Escherichia coli ssuE: defining a general catalytic cycle for FMN reductases of the flavodoxin-like superfamily[J]. Biochemistry, 2014, 53(21):3509-3519.
Eichhorn E, van der Ploeg J R, Leisinger T. Characterization of a two-component alkanesulfonate monooxygenase from Escherichia coli [J]. J Biol Chem, 1999, 274(38):26639-26646.
van Der Ploeg J R, Iwanicka-Nowicka R, Bykowski T, et al. The Escherichia coli ssuEADCB gene cluster is required for the utilization of sulfur from aliphatic sulfonates and is regulated by the transcriptional activator Cbl[J]. J Biol Chem, 1999, 274(41):29358-29365.
Borkowski O, Bricio C, Murgiano M, et al. Cell-free prediction of protein expression costs for growing cells[J]. Nature Communications, 2018, 9(1):1457-1457.
Lian S, Diko C S, Yan Y, et al. Characterization of biogenic selenium nanoparticles derived from cell-free extracts of a novel yeast Magnusiomyces ingens [J]. 3 Biotech, 2019, 9(6):221-221.