Team:Exeter/Design
Reactions and Enzymes
Our goal was to produce oxygen from perchlorate, which is a two-step reaction shown in Figure 1 and 2. There are dissimilatory perchlorate reducing bacteria (DPRB) that can perform these reactions naturally, but they have very particular growth conditions. We decided to use E. coli because it is a non-fastidious organism that is well researched. Additionally, E. coli has existing mechanisms, such as a TAT transport system and available electrons in the quinone pool, that we planned to harness.The DPRB in nature produce five proteins that makes perchlorate reduction possible. Four of them form the perchlorate reductase complex, catalysing the reaction in Figure 1, and one is chlorite dismutase, catalysing the reaction in Figure 2. Perchlorate reductase is encoded by the pcrABCD operon and chlorite dismutase is encoded by the cld gene.
Genetic Parts
The pcrABCD operon has four subunits. Of those, pcrA and pcrB produce enzymes that combine into perchlorate reductase, while pcrC and pcrD produce accessory proteins. PcrC supplies electrons from the quinone pool and PcrD supplies a molybdenum cofacter to pcrA, and potentially plays a role in the fusion of smaller proteins produced by pcrA and pcrB (Bender et al, 2005). pcrABCD is quite a large operon (with complete operon size 5555 bp), so there was some concern about being able to synthesize it to begin with, especially as this is where previous iGEM teams struggled. There was some hope of being able to remove pcrC and instead supply electrons by applying an electric field across the bacteria, but after talking to Professor Mike Allen, one of our primary stakeholders, it was decided that electricity would be too much of a risk, due to to perchlorates high oxidation power. In the end, we had to split up the operon into pcrA and pcrBCD genes, then recombine them using biobrick assembly.
Because pcrABCD and cld produces large proteins that cause a significant metabolic burden on the bacteria, the inducible T7 promoter was used. This ensured the bacteria could grow without this metabolic burden, and protein production could be induced when ready. We used the B0034 RBS and the B0015 terminator because they are strong and well characterized, making them standard parts in iGEM projects.
In DPRB, Cld, PcrA, and PcrC are transported to the periplasm using unique signal peptides and the bacteria's TAT transport system. We wanted to harness the E. coli TAT system, so we had to confirm that the signal pepides would interact wi the E. coli TAT system in the same way. We did this by attaching GFP to these signal peptides, self fractionating the cell and then performing a Western blot, to see if GFP was in the periplasm. You can read more about this experiment here.
Modelling
Different DPRB contain different versions of Cld and Pcr, which have different enyzme kinetics, and we have explored the possibility of this using modelling. Because of the pcrABCD operon's size, we could only synthesize one version of it, so we chose the Pcr from the model organism of perchlorate reducing bacteria, A. Suillium. However, for the smaller Cld, we could afford to synthesize different versions. From the literature values (Hofbauer et.al, 2014) of the rates of reaction, we modelled which Cld's would work with our Pcr. It was found that the Nitrospira defluvii sourced chlorite dismutase (NdCld) coding sequence would not react fast enough with chlorite, leading to a lethal buildup. The Dechloromonas aromatica Cld (DaCld) was found to be optimum while still reacting fast enough to prevent buildup.Our modelling also informed many of our decisions about our bioreactor design, such as substrate concentration and required size. We found that a bioreactor of 48L in volume was needed in order to sustain a living person with the predicted rates of oxygen production, and were able to obtain a model bioreactor of 1:10 scale.
You can read more about our model here.
Applications
To keep the bacteria contained, protect them from UV radiation on Mars, and harvest the produced oxygen we knew a bioreactor was necessary. Our bioreactor was where our human practices was the most relevent, as we interviewed many stakeholders to inform our designs and to intergrate them into existing infrastructure. You can look at our design timeline here.To protect our bacteria, we decided our bioreactor should be covered in high-Z steel-steel composite metal foam, which is a lightweight material that blocks radiation. With input from some stakeholders it was decided that a swirl flow type bioreactor should be used, since it can both mix the perchlorate with the E. coli, and sort oxygen from the water and the bacteria. The bacteria themselves would be contained within alginate beads which would be flowed through the bioreactor. We continued to work with stakeholders to optimize our bioreactor Mars, and to find ways to integrate it into existing life support systems that could potentially be used on the Red Planet. Our stakeholder from NASA, Melanie Pickett, specifically asked for a CAD diagram, which can be seen in Figure 7.
Summary
Over the course of this project, our aims were to insert and express the genes encoding for perchlorate reductase and chlorite dismutase. We also wanted to test the signal peptides responsible for transporting these enzymes in the periplasm. In order to choose which bacteria we wanted to obtain our genes from we used a model which also informed the parameters of our bioreactor design. We wanted to design a bioreactor specifically for producing oxygen on the Martian surface, and interviewed many experts contributing towards the effort to get to Mars in order to do this. Though there is still much to do, we managed to accomplish all these goals during our project.
References:
- Bender, K.S. et.al, (2005). Identification,characterisation, and classification of genes in coding perchlorate reductase J. Bacterial. 187. 2 5090-5096
- Kostan, Julius, et. al,. (2010) “Structural and Functional Characterisation of the Chlorite Dismutase from the Nitrite-Oxidizing Bacterium ‘Candidatus Nitrospira Defluvii'” Journal of Structural Biology, 172
Back to Top