Team:GreatBay SZ/Proof Of Concept
Overview
Our story began with the Internet of Things, or IoT. Basically, it refers to the billions of physical devices that are collecting and sharing data around the world. The aim of GreatBay_SZ 2020 is to develop BIOT, an electric generator harvesting energy from moisture in the air to solve the power-supply issues billions of IoT devices are facing.
If you want to know more about the background story of BIOT, please read our Description Page.Our project is mainly divided into four sections - BIOT production; Cost reducing; Power Efficiency Improvement and Hardware design. In all sections, we have achieved promising results. Next, we will introduce each part in detail.
BIOT Production
Genetic Circuits Design for nanowire production
The Core of our BIOT is Electrically-conductive Protein Nanowires (e-PNs). They are originally produced from G. sulfurreducens. But barriers to large-scale e-PN production have been a limitation to realizing the potential of Geobacter e-PNs. Geobacter sulfurreducens must grow anaerobically, which adds technical complexity and costs. The limiting options for controlling the timing and extent of e-PN expression is another major barrier in this non-model microorganism.
We thought that Escherichia coli might be an ideal chassis for e-PN fabrication. E. coli is a common platform for the commercial-scale production of organic commodities. There are also rich tool boxes for tuning gene expression in E. coli[1].
Figure 1: The difference between E. coli and Geobacter sulfurreducens.
To express BIOT nanowires in E. coli, through literature research and many academic discussions, we designed the below genetic circuits[1]. It is mainly composed of two parts:
Figure 2: The X-ray Structure of e-PN monomer protein (a) and Western blot to verify the expression of BIOT monomers (b).
The first part is the protein nanowire monomer. We fused a PpdD signal peptide (MDKQRG) to the N-terminus of the protein monomer, which can help the protein to be secreted outside the cell membrane. We verified that BIOT's protein monomers were successfully expressed in E. coli using His antibody against the C-terminus His tag of monomer protein in western blot.
Figure 3: The Genetic circuit of BIOT.
The second part is the Type IV secretion systemn composed of 12 genes. It is a secreted protein machine distributed on the cell membrane, which is used to assemble BIOT's protein monomers into nanowires that can be up to 20 microns in length.
Table 1: The composition of BIOT Genetic Circuits.
Figure 4: The scheme of the BIOT nanowires on the E. coli (a) and microscopy of the Type IV nanowires produced by E. coli[2] (b). BIOT nanowires are supposed to have a similar complex structure.
Before transferring the Type IV secretion system to E. coli Top10, we had to remove the Type I system. If the two systems were all present, the assembly efficiency of e-PNs would be affected. FimA was the key gene in the original Type I system, so we deleted the gene fimA in E. coli top10 using CRISPR-Cas9 system to prevent the formation of type I pili. From the graph, we can see that the genome fragment without fimA is shorter than the one from the wild-type strain, which means the success of fimA gene knockout.
Figure 5: CRISPR-Cas9 system[3] for the deletion of fimA gene in E. coli Top10.
Figure 6: The gRNA sequence used in our project (a) and the PCR verification of the deletion of fimA in E. coli Top10 using CRISPR-Cas9 system (b).
Finally, we introduced the constructed expression plasmid into the fimA knockout E. coli, and the BIOT nanowires can be successfully expressed on the membrane.
BIOT purification
Then our next task is how to obtain and purify the nanowires from the cell membrane. The whole procedure for BIOT purification included cell cultivation, nanowire blending, nanowire filtration and concentration measurement.
The final concentration of protein solution was roughly around 200 to 400 ng/μl.
Figure 7: The procedures for BIOT nanowires purification. (a) Cell culture; (b) Nanowires blending; (c) Nanowires filtration; (d) Protein concentration measurement.
Figure 8: BIOT nanowired was filtrated on a 100kDa ultrafiltration membrane (a) and the Protein Concentration of BIOT purification (b).
You can see the nice protocol video we make below to have an overview of the whole purification process.
BIOT Battery Fabrication
As you can see from the picture, it is the basic module of BIOT. The positive gold electrode at the bottom is larger so that the e-PNs can be positioned on it. And the tiny circular carbon piece works as a negative electrode on the top.
Figure 9: BIOT Assembly and Measurement. (a) The basic parameter of the basic BIOT module; (b) Different ways for BIOT electric generator assembly; (c) Digiital multimeter, Keithley 2401 was used for Voltage and Current measurement.
Under 60% relative humidity, we obtained a stable output of roughly 0.35V from a single module! The current fluctuated from 0.2μA to 2μA. Additionally, to find the optimum relative humidity for the battery to function, we did the measurement with different relative humidities. The optimum RH was roughly at 40-50%, where the voltage rises to about 0.42V.
Figure 10: The Key parameters of BIOT basic module. (a) Voltage; (b) Current; (c) Voltage under different humidity conditions;
Video showed the Voltage and Current were detected from BIOT electricity generator. (Different time)
After arduous experiment lasting for nearly two months, it is not easy to get to this point. When we finally produced electricity, we were very excited. The CCTV camera in the laboratory recorded this exciting moment.
In short, we designed a gene circuit to express nanowires and successfully verified its expression in E. coli. We made a battery based on the nanowires and successfully got the power from BIOT by harvesting the moisture power in the air.
We've successfully realized the power generation from the air, but that's not enough. What we aimed at, was to make a difference and release the full potential of BIOT.
Cost Reducing
High-efficient production of protein nanowire for large scale application
First, Cost Matters!
The important reason why we choose E. coli to express nanowires is its abundant regulatory tools. It's easy to know that the Type IV system on the member could be one of the limiting processes in the nanowires assembly.
We then take the RBS optimization strategy to target the RBSs of the Type IV e-PNs assembly system. We expect this strategy[4] can increase the number of type IV secretion machines on the membrane, thus optimizing the production of the e-PNs, then lowering the cost. As shown in the figure below, there are mainly three RBSs, which are RBS of hofB, hofM and ppdA. These three RBS sites have been reported in the literature that their sequence changes can optimize the expression of nanowires in E. coli.
We originally planned to create a library for each RBS, then randomly combine them to find the optimum combination. So we used the RBS Calculator from Salis Lab, obtaining 9 more mutated sequences with various strengths for each. But, with ten sequences for each of the 3 RBSs, there're 1000 possibilities! Therefore we would need to spend 250 days on the purification process (4 purifications/Day). It is impossible in one-year iGEM Competition.
Figure 11: Building RBS library to Improve the production of BIOT. (a) Three RBS was selected for library construction; (b) The sequence and Translation initial strength (T.I.R) of wildtype RBS for three genes; (c) The scheme of the library size and strength of three RBS;
Thus, we were forced to reduce the size of our RBS library down to 3*2*3. Due to time constrain, we only managed to successfully introduce six combinations out of 18 into the △fimA strain, but we still obtained some optimistic results at the end. As the table and the graph shows, with the optimum mutation, RBS4, productivity was successfully increased by 34.1%. You can see the RBS information for each mutant in the below figure. In this part, we have achieved a further increase in BIOT production, thereby reducing its costs and promoting its further application in the IoT field.
Figure 12: The 3*2*3 RBS library combinations (a) and the sequence of each designed RBS (b) .
Figure 12-1: RBS optimization increase the production of BIOT (a) and the RBS combinations for each optimized mutant.
Power Efficiency Improvement
Protein engineering to improve power efficiency
Second，Power Efficiency Matters!
Here, in this section, we were aiming to apply protein engineering technology based on two strategies to modify the protein monomer of e-PN, hoping to increase its voltage level.
1. Increasing the number of carboxyl groups on nanowires.
One of the perspectives we considered was the number of carboxyl groups on nanowires.
We established mathematics models to explore the relationship between the number of carboxylic groups and the output voltage. As you can see in the Model Page, when the humidity is at the same level, the potential value is proportional to the number of carboxyl groups. By increasing the number of carboxyl groups on the e-PN monomer, we can theoretically improve the potential of the basic BIOT module.
Figure 13: The model for BIOT potential generation.
To verify our idea, we designed 16 mutants of the e-PN gene, to increase the number of carboxyl groups which reacted with water molecules to form hydrogen ions. We made it by replacing 1 to 4 amino acids in non-conservative regions with Aspartic acid or Glutamic acid. That's because these two amino acids both have two carboxyl groups. The amino acids we chose to replace had similar structures to them based on the Grantham's distance and other methodologies, according to our research.
Figure 14: Aspartic acid (Asp) and Glutamic acid (Glu) both have extra -COOH.
Figure 15: Figure 14: Increase the potential value of BIOT by adding more Carboxylic groups. (a) The Grantham's distance model for amino acid similarity; (b) The sites (green) selected for adding -COOH into the BIOT monomers;
You can see the mutated site of mutants with 1-4 carboxylic groups added below.
Table 2: The mutant sites of mutants with 1-4 Carboxylic groups added.
2. Decreasing the diameter of the nanowires.
Another design we considered was to decrease the diameter of the nanowires. By decreasing the diameter, it becomes harder for the water molecules to pass through the nanopores; it will lead to greater water potential gradient as reported from the work published on Nature.
Through literature search, we found two mutations F51W and Y57W can make the diameter of nanowires decrease from 3nm to 1.5nm.
Figure 16: Increase the potential value of BIOT by decreasing the diameter of nanowires.
These mutations mean changes to the structure. Due to the huge workload of determining the performance of each mutant, We first used Swiss Model to predict the structure of every mutant to examine whether the mutations affect the structure dramatically and then analyze the Root-mean-square deviation of atomic positions, RMSD, between wild-type and mutants using PyMOL. The smaller the value is, the more similar they are. The results were all below 0.03, far lower from the threshold, which confirmed that our design would work successfully.
Figure 17: RMSD value of the mutants compared to the wildtype nanowires.
Then we clone, transform and purify of the mutants and test their performance. From the chart of the output voltage of different mutated proteins, we can see that the 4A mutant gave a rather ideal result, with the voltage increased by 46%. It reached 0.51V compared to the original 0.35V! It's definitely an essential step in our project. The output of the thinner mutant THIN was 0.47V, which also increased from the original voltage by 34.3%. It means that the two designs both work as expected.
Figure 18: The output voltage of different mutants.
Since there's an obvious increasing trend from 1A to 4A & THIN, we did a repetitive measurement with these five mutated proteins. Surprisingly, the difference was narrowed (Figure 18b), but the result was enhanced generally. We admit that the output voltage of the mutated protein is still unstable, and the reason behind it needs further analysis. But anyway, it provides a new idea for future study.
The analysis of nearly 25 mutants in the above RBS Optimization and Protein Engineering sections is a huge workload, including molecular cloning, protein purification, and battery fabrication. The following picture shows the number of plates we used in this summer. They are about 700-1000.
Figure 19: All the plates GreatBay_SZ used this summer.
Explore Full Potential
Design hardware to harness the power of the ambient environment
To test whether our BIOT could be utilized in reality, we did some further design on the hardware.
1. Connect more, achieve higher.
We firstly connected more than one basic BIOT modules in series.
Figure 20: Voltage output from connected nanowire devices in series.
At first, we made a board that can be used to connect multiple batteries in series, with the designed copper sheet structure on top of it. Using this board, we can easily connect multiple batteries in series. As seen in the video，and with four of them, we managed to increase the output up to 1.6V. It proved that it could work, which means it is going to work with five, ten and even more of them, etc.
Figure 20: (a & b) The board with the designed copper sheet structure on top of it; (c) Connected nanowire devices in series lead to higher voltage output.
Connected 4 nanowire devices in series lead to 1.6V output.
We then calculated the theoretical output power of the BIOT connected in series through a simple mathematical model, and the value can reach a maximum value of 1KW/m3. This value is theoretically higher than the existing solar cell based on solar power（typical power:100W/m2）, because the drawback of solar cells is that, unlike our BIOT, they cannot be superimposed on the vertical axis, which limits the maximum output power of solar cells.
Figure 21: The theoretical output power of the BIOT compared to solar cell.
2. Stable the output for more scenarios.
In order to better stabilize the output power of BIOT, we decided to add a PMIC module into the circuit.
Power management integrated circuits (PMIC) are integrated circuits for power management. The environmental energy itself has the characteristics of unstable power output, and the PMIC module specially developed for environmental energy technology (such as BQ25570 or LTC3588) can store and manage the energy from the environment, and output electrical energy in a more stable way.
Figure 22: Using PMIC module to stable the BIOT output.
In our design, PMIC will temporarily store BIOT's output power through its power management chip and power the IOT devices stably. The PMIC module will further enable BIOT to meet the needs of more scenarios. Below is an example to power the LED using PMIC.
Figure 23: Using PMIC module to power the LED.
Using PMIC module to power the LED.
In summary
The scenarios for the future
With its omnipotent power, we can expect IoT to play a major role in the future, from maintaining the optimum humidity in a house, to impeding the progress of global warming. In the foreseeable 2030, there will be approximately 125 billion IoT devices; every second, 127 devices are being connected to the network.
However, as you can see from the Description, the Power supply is always a critical issue for IoT devices, especially for those who cannot easily reach the traditional power. The electric generator, BIOT, that utilizes a thin film of protein nanowires to harvest energy from moisture in the air can offer a strong solution to this big issue.
Throughout the whole summer, our team have successfully designed this high-performance electric generator that's driven by moisture in the air and increased its productivity and power efficiency using synthetic biology. As far as we know, this will be the first time that biotechnology has been applied in the field of IoT.
We envisioned a Promising, Optimistic and Bright future of BIOT. With BIOT, we can see a grand future where hundreds of billions of IoT devices connected the whole world.
References
[1]Liu,X., Gao,H., Ward,J.E., Liu,X., Yin,B., Fu,T., Chen,J., Lovley,D.R. and Yao,J. (2020) Power generation from ambient humidity using protein nanowires. Nature, 578, 1–5.
[2] Liu,X., Gao,H., Ward,J.E., Liu,X., Yin,B., Fu,T., Chen,J., Lovley,D.R. and Yao,J. (2020) Power generation from ambient humidity using protein nanowires. Nature, 578, 1–5.
[3] Jiang,Y., Chen,B., Duan,C., Sun,B., Yang,J. and Yang,S. (2015) Multigene editing in the Escherichia coli genome via the CRISPR-Cas9 system. Appl Environ Microb, 81, 2506–14.
[4] Li,T., Ye,J., Shen,R., Zong,Y., Zhao,X., Lou,C. and Chen,G.-Q. (2016) Semirational Approach for Ultrahigh Poly(3-hydroxybutyrate) Accumulation in Escherichia coli by Combining One-Step Library Construction and High-Throughput Screening. ACS Synthetic Biology, 5, acssynbio.6b00083..