Propane pathway

Background

Microbially produced propane holds enormous promise as a potential replacement of portable fossil fuels, but the propane yields with current biological pathways are low. Our pathway starts from acetyl-CoA and produces propane through 6 intermediates. To help concentrate engineering efforts on its critical parts, better quantitative understanding of the pathway is required. Our goals were to build a mathematical model of the pathway to better understand it and create and test BioBricks of the propane pathway to help future teams and researchers to continue improving it.

Outcome

• Built a model of the pathway based on known kinetic properties of the enzymes

• Identified major bottlenecks of the propane pathway using our model

• Improved our experimental plans according to the modeling results by changing one enzyme to a better homolog and expressing the rate-limiting enzyme from the highest copy number backbone used

• Found that propane output was sensitive to NADPH/NADH, suggesting their efficient regeneration might be a limiting factor

• Designed two BioBricks containing all ten genes required for the propane pathway to produce propane in E. coli

  • Chose three intercompatible backbones to ensure that the propane pathway could be integrated to the same cell with cellulose hydrolysis (which required the third backbone)





• Used NEBuilder (Gibson) Assembly to construct the BioBricks

  • Submitted a BioBrick (Propane 1) containing three crucial enzymes of the propane pathway and inserted it in pSB6A1 backbone for usage in propane production

  • After trying both ELIC and OE-PCR once and Gibson assembly twice, we managed to assemble Propane 2 with Gibson assembly and detect the right size insert with colony PCR. However, as we ran out of time, we were unable to successfully propagate the correct plasmid, despite multiple tries. As we didn’t manage to assemble the whole pathway, we couldn’t try out whether the pathway was functional.

Continuous production

Background

Industrial biopropane production would most likely occur as a continuously operated process, as it is economically more feasible in large scale production. To take the first step towards industrial scale production we wanted to try continuous production of propane using a E. coli strain provided to us by Pauli Kallio from the University of Turku.

Outcome

• Successful batch production to prepare analytical equipment for continuous production

• First report ever worldwide of successful microbial production of propane in continuous production

• 145 hour continuous production experiment in a 500 ml chemostat

• Propane yield up to 22.7 µg/L in reactor gas phase

Cellulose degradation

Background

Well over 200 million tonnes of cellulosic waste is left unused each year in the European Union alone. To elevate the microbially produced propane to a 2nd generation biofuel and avoid interfering with food production, we wanted to incorporate a plasmid for cellulose hydrolysis into the same bacteria that produces the propane. The plasmid contains three genes encoding the enzymes that hydrolyse cellulose polymers into glucose.

Outcome

• Looked into modeling cellulose breakdown, but found that there was not enough information to model the breakdown sufficiently well to get any practical benefit from the model.

• After trying Gibson assembly and ELIC twice, we were able to produce the Cellulose insert containing the genes encoding three enzymes required for cellulose hydrolysis with OE-PCR. However, due to its low concentration, we were unable to transfer it to a backbone and propagate it.

Amphiphilic protein

Background

Amphiphilic proteins are synthetic proteins consisting of a hydrophilic and a hydrophobic domain that have been shown to spontaneously form micellar and vesicular structures. We were interested whether these structures could be used as scaffolds to have subsequent enzymes of the propane pathway in close proximity. We wanted to model whether fusing enzymes to the proteins would disrupt micelle formation and whether our idea could enhance propane output.

Outcome

• Built a stochastic synergy model in Python, simulating enzyme function in cases where two subsequent enzymes stay in close proximity to each other as opposed to moving around independently in a cell

• The synergy model predicts a 200-400 % increase in product output if enzymes stay in close proximity to each other

• Constructed a geometrical micelle model based on the sizes and structures of the micelle-forming proteins, indicating that it is indeed possible for micellar structures to form even as enzymes are fused to them

• Submitted a BioBrick unlikely encoding the amphiphilic protein to the registry

• Due to time restraints, we were unable to experimentally validate the idea by fusing either two subsequent enzymes of the propane pathway or components of the violacein pathway to the amphiphilic proteins

N-terminally fusable GFP

Background

To validate our amphiphilic brick, we needed a GFP that could be fused to the amino-terminal end of the protein. There was no such brick in the distribution kit. We also thought there was no such brick in the registry, but later came to realize we were wrong. We wanted to create a GFP BioBrick that could be fused to the N-terminal end of any protein using BioBrick methods.

Outcome

• Submitted a BioBrick encoding GFP with an extra nucleotide prior to the suffix, ensuring that it can be fused to the N-terminal end of a protein using BioBrick enzyme assembly methods while maintaining the reading frame

• Collaborated with team HS Slovenia to validate the brick

Combining modeling and experimental work in iGEM

Background

Mathematical modeling is a key component of synthetic biology and also played a central role in our project. Collaboration between modelers and biologists can however be challenging, something we also noticed in our project. We wanted to study how iGEM teams tackle these challenges and are able to integrate modeling to their experimentation.

Outcome

• Created a questionnaire for iGEM teams on collaboration between modeling and experimentation, and studied 2014 teams and professional synthetic biology groups to find out what educational backgrounds team members are coming from

• Biggest issues in collaboration between modelers and experimentalists are: lack of knowledge of the other field, lack of common terminology and differences in ways of thinking

• Both modelers and biologists need to understand the basics of the other field to be able to effectively collaborate.

• Having experimentalists and modelers work close together is beneficial. One approach generally found successful is to have some biologists get involved in modeling to help ensure models are useful for the project and connected to reality. Modelers can too take some time to get familiar with the wetlab in the beginning of the project

• Regular team meetings for presenting and discussing progress and issues of every field take time, but ensure all team members stay informed and can voice their insights.

• Students with a mathematical background are underrepresented in iGEM teams as compared to professional synthetic biology groups.

• iGEM teams have relatively many biotechnology students, who often stand seem to act as mediators between the modeling and experimentation.

Software: HumHub report and Collab Seeker

Background

Finding new collaboration partners in iGEM is not easy, as finding information about different teams' projects is time-consuming and often difficult. On the other hand, communication with other teams as well as internal team communication can be difficult due to a multitude of platforms used, often cluttered with non-iGEM content. We wanted to do something to these issues to make iGEM even better.

Outcome

• Worked with Stockholm iGEM team on HumHub, a collaboration platform for iGEM teams, and collaboratively wrote a report on it with them

• Created a questionnaire on how teams found their collaboration partners and how they’re keeping in touch with them

• Found that 22 out of the 23 teams that answered wished for better means to find collaboration partners with

• Built Collab Seeker, a lightweight collaboration search tool, which helps find relevant collaboration partners using keywords and provides their contact information

Future

To read our thoughts on future prospects and on how to carry on from where we left, please see our Future page.

Submitted BioBricks and Achievements

To read about the BioBricks we submitted, see our Submitted Parts page. To see how our achievements match with the medal criteria, see our Checklist.