Results

Integrated Human Practises

Goal: We aimed to develop meaningful connections with stakeholders in all parts of the produce packaging pipeline, and to use their expertise to shape our project design into a solution that reflects the needs and values of those it will impact.

Result: The feedback and perspectives of industry and community stakeholders have been integral to our project, and have shaped every iteration of our design. Their input informed multiple key project pivots, including the transition from a wrapping to a box, the type and integration approach of our antimicrobial peptide, and the abandonment of a proposed detection system. Our HP contacts also gave important technical recommendations, such as how to conduct antimicrobial tests and identify microbe types. The conception and refinement of our PHB subproject was also thanks to HP input. Finally, and perhaps most importantly, our stakeholders validated the need for our proposed solution, which is critical to a responsible and effective project.

Future Directions: While the current design of our project has been shaped by integrated feedback from these stakeholders, the need for community input does not end here. As we develop further iterations of Cellucoat, our aim is to connect with more retailers and distributors to test and validate our solution at the industrial level. Importantly, we plan to enter conversations with the International Fresh Produce Association to explore the potential for Cellucoat to meet emerging needs for sustainable packaging solutions. To develop production feasibility, a future direction includes testing the integration of our antimicrobial material with current production systems. Finally, our aim moving forward would be to integrate more feedback about the prototype design of Cellucoat at both the consumer and end-user level, with regards to factors such as color, shape and size.

Nisin Expression and Characterization

Goal: Nisin is an antimicrobial peptide consisting of 34 amino acids. It is a food safe and thermo stable protein which is currently being used as a preservative in canned foods. Our goal was to incorporate nisin Q (NisQ) into our bacterial cellulose using a co-culture. We designed multiple gene fragments with varying primer cut sites to be inserted in varying plasmid backbones. This plasmid with the nisin gene fragment inserted into it would be transformed into the BL21 strain of E. coli.

Result: We were able to insert our gene fragment 1 which consists of XbaI and EcoRI cut sites. This was successfully cloned and inserted into a pSB1A3 backbone. A sequencing of our pSB1A3-nisin plasmid was sent in to analyze how well of a match our inserted sequence is in comparison to our original nisin sequence. We had accurate sequencing results indicating successful cloning. We also cloned and transformed E. coli BL21 cells with our GST-NusA-NisQ-His composite part and successfully produced protein, as indicated by our SDS-PAGE and Western gel results. The GST solubility tag increased our NisQ protein size from 7kDa to 91kDa, making it easier to visualize on a gel.

To characterize our antimicrobial peptide we also ran Kirby Bauer disc diffusion tests against B. subtilis using co-cultured bacterial cellulose paper discs immobilized with purchases nisin. The tests showed that nisin can inhibit bacterial growth for up to 6 hours at optimal conditions.

Future Directions: To further our nisin characterization and production results, our team will purify NisQ from the GST and NusA solubility tags for Ni-NTA protein purification and Kirby-Bauer tests against B. subtilis. Once successful protein expression has been accomplished, we will use our strain of BL21 E. coli which can express nisin, in a co-culture with K. xylinus where it will begin to produce protein and functionalize the bacterial cellulose.

PHB Improvement and Expression

Goal: We aimed to improve the strength and durability of our bacterial cellulose (BC) product by fortifying it with polyhydroxybutyrate, a biodegradable polymer produced in bacteria under nutrient-limiting conditions. Using parts designed by previous iGEM teams, we designed a recombinant PHB-production and secretion system in E. coli with a variable expression system of phasin, an associated protein which plays a role in secretion. Ideally, by modifying phasin expression levels with variable ribosomal binding site strengths, the resulting system can be integrated into our co-culture model to produce a PHB-BC composite with fine-tunable mechanical properties.

Result:The purpose of experimenting with different strengths of RBS upstream of the phasin-hlyA gene was to create a set of constructs that can be used to fine-tune PHB expression and secretion by up or down-regulating phasin translation. Our Western blot results indicate that the stronger the RBS upstream of the phasin-hlyA gene will upregulate the amount of phasin produced.

Future Directions: Having successfully modified phasin expression levels using ribosomal binding strengths, and subsequent PHB secretion levels, this subproject represents a promising tool for modifying relative concentrations of PHB incorporated into a PHB-BC composite material. As PHB has an effect on BC’s mechanical properties, including flexibility and tear resistance, this represents a means to meet a variety of packaging material needs beyond hard plastic. As such, future directions include testing the effect of varying PHB concentrations on PHB-BC mechanical properties, and validating the applications of the materials in the packaging industry.

Co-culture

Goal: Bacterial cellulose (BC) is a promising material because of its potential to be functionalized. A method to functionalize BC would be to integrate the molecules simultaneously as the BC is being produced through a co-culture. Multiple factors come into play when generating a co-culture, so this subproject focuses on how to both optimize co-culture conditions for BC and recombinant protein production and post-production treatments of the BC material.

Result: To determine the optimal media to grow the E. coli and K. xylinus monocultures, the results indicate that using either HS media or HS media enriched with tryptone will have similar BC yields. Next, assessing the impact of K. xylinus and E. coli extracellular secretions on the other organism’s growth indicated that the extracellular secretions of E. coli increase BC yield by 51.61% and have no significant impact on E. coli cell density. Lastly, the final BC yield from the co-culture is 27.05% lower than the BC yield from a monoculture.

In regards to the post-production treatments, the results indicate that a four-day 0.5 M sodium bicarbonate purification treatment, 20-minute boiling water bath, and leave to air dry for two to three days results in a material with the greatest transparency, homogeneity in appearance, and retained strength compared to a 0.125 M NaOH bath.

Future Directions: Future experimentation with these two strains of E. coli and K. xylinus in a co-culture will help determine how byproducts of each population’s growth impacts the growth of the other populations, and how each changes the properties of the final BC product. Furthermore, if the co-culture would be used on an industrial scale to produce BC, PHB, and the antimicrobial peptide for weeks at a time, antibiotics must be added to the co-culture to create a selective environment. Hence, K. xylinus will be given antibiotic resistance to chloramphenicol with the IBMc396 plasmid.

Golden Gate Assembly

Goal: The drawback to using nisin is its narrow spectrum of antimicrobial activity, as seen through how nisin is a potent agent only against certain fungi and gram-positive bacteria.To give our stakeholders in industry the option to customize their antimicrbial peptide, this subproject aims to insert nisin into our RFP flipper expression vector through Golden Gate assembly.

Result: We were able to successfully insert our nisin insert into our RFP flipper, however, were unable to complete sequencing our construct to confirm that the nisin insert was added.

Future Directions: After being able to insert nisin into our RFP flipper device, next steps include swapping nisin with one or more antimicrobial peptide inserts (done through overlapping overhangs). Candidates for potential antimicrbial peptides to further this proof of concept include Iturin A, Bac5, and Pumilacidin.

FWM

Goal: The conventional K. xylinus growth media, Hestrinn-Schramm (HS) media, represents up to 65% of BC production costs. Given the comparatively low costs of plastic, this represents a key focus for efforts to increase the economic viability of our proposed solution. Inspired by Julianna Schneider and a scoping of current literature, the goal of this subproject is to supplement HS media with enzymatically hydrolysed orange peel. If successful, this serves as a proof-of-concept for the use of other fruit waste in BC production.

Result: Based on the successful growth of BC in a co-culture supplemented with our fruit waste media (FWM), we were able to show that FWM can sufficiently replace part of the HS media. In our competitive cost analysis, we demonstrate that the 45% pulp mixture was the most cost-effective FWM composition, which reduced BC production cost by a staggering 60%. Significantly, uniaxial tests confirmed that this cost reduction did not compromise the mechanical strength of the BC product. In fact, the supplementation of HS with some FWM compositions resulted in increased strength: compared to the sample grown in 100% HS media, the 30% juice sample and the 45% pulp samples showed 2.5 and 1.4 times increase in strength, respectively.

Future Directions: This subproject successfully demonstrated that fruit waste media made from navel oranges is a viable, and even beneficial, supplement to HS media. For future production initiatives, this concept can be expanded to include other types of fruit waste, which are prolific in the produce industry. Future experiments could also incorporate a wider range of enzymes for a more complete breakdown of the waste components into glucose.

Entrepreneurship

Goal: The entrepreneurship subgroup aimed to build a business case for commercializing Cellucoat as it takes the next step to becoming a venture.

Result: We conducted and analyzed customer discovery interviews for Cellucoat in the development of a business plan, cost analysis, and a discourse around the implications and future of Cellucoat through a legal, ethical, and social analysis.

Future Directions: Before Cellucoat can become a widely used product, two key areas need to first be developed: 1. The ability to reduce the cost of production and 2. Optimizing the growth conditions of both K. xylinus and E. coli, we can maximize the BC yield. To aid in the development of this, we plan to keep refining the entrepreneurship of Cellucoat through pitching competitions and launchpad programs.

Education and Public Engagement

Goal: We aimed to introduce synthetic biology to grade school students across Calgary through presentations and demonstrations. We wanted to introduce them to the basic concepts of genetic engineering, spark interest in research and synthetic biology, and engage students in relevant hands-on activities.

Result: We visited three schools across Calgary, engaging with approximately 250 students. We received feedback from over 150 students and educators that indicated these presentations were effective and engaging. 80.8% of respondents found the workshop enjoyable, and 58.7% felt that the workshop increased their understanding of synthetic biology. 78.4% of respondents expressed potential interest in a future synthetic biology class integrated into their curriculum.

Future Directions: After receiving overwhelmingly positive feedback, we adapted our presentations into a toolkit for educators looking to introduce synthetic biology to their students. This included a video introduction to synthetic biology, a video for educators explaining how to do the demo we performed, and a video for students demonstrating the demo. The toolkit also included additional resources and FAQs for teachers. We plan to disseminate this toolkit to educators at the schools we visited, and continue gathering feedback about how this toolkit can be bettered and expanded.

Involutional Neural Network

Goal: We sought to replicate the AMPScanner, the CNN used by Veltri et al. as an Involutional Neural Network (INN). INNs approach their data in a spatial-specific, channel-agnostic way, and have a lower computational cost than CNNs on generic image datasets while maintaining equal or greater accuracy. Our replication aimed to investigate whether INNs perform comparably to traditional CNNs on biological data (FASTA sequences).

Result: Our INN replication performed comparable to the Veltri et al. CNN prior to hyperparameter tuning, using measures of sensitivity, specificity, and accuracy. (Table of measures). After hyperparameter tuning, the accuracy of the INN remained the same.

Future Directions: Hyperparameter tuning did not result in a notable increase in performance for the INN. In the future, it would be valuable to test whether different approaches to tuning, or a different method of fitting our data to our INN, would result in performance increases. As we had to make significant changes to the Veltri et al. paper in order to run the original CNN, and the original paper did not report measures of computational load, we were not able to determine whether INNs perform better in this regard. Fully recreating Veltri et al. using the same versions of software as our INN in the future will enable us to accurately compare measures of computational load (such as computation time and memory use).

As the INN still performed as accurately as the conventional CNN, it may be used in the future combined with Golden Gate in order to modularize Cellucoat; the INN can be used to identify peptide sequences that are antimicrobial, and can be expressed using Golden Gate.

BioSculpting

Goal: We aimed to develop and iterate a production method for clamshell-like BC packaging that incorporated many of the strengths of the material while mitigating many of its shortcomings. The project also aimed to find ways to mechanically strengthen BC without the incorporation of other materials.

Result: The subproject resulted in the creation of multiple BC box prototypes, ranging from small, 64 cm³ boxes to large 720 cm³ boxes. We also developed a workflow and the appropriate hardware to produce BC cardboard, allowing for our unmodified BC to gain crucial mechanical strength.

Future Directions: Naturally, our future directions will be influenced towards the creation of larger prototypes to test our workflow on larger scales. We aim to further iterate the development of BC cardboard by testing more effective and efficient methods of production, hoping to mimic the rapid production found in other packaging materials. We also aim to reduce the amount of wasted BC used in the process, finding alternative uses for scrap pieces and, if not possible, incorporating the wasted BC into compositing workflows.

SAMARA

Goal: SAMARA aimed to serve as a comprehensive, modular replacement for 2018 Calgary’s SARA subproject. Specifically, we aimed to solve SARA’s issue of extensive manual summarization while simultaneously providing a comprehensive scraping framework for future teams to build off of.

Result: We successfully produced SAMARA in two components. First, the iGEMScraper, a Scrapy CrawlSpider designed to crawl through iGEM wiki pages, parse the data, filter and clean it, and then pass it to a summarizer to extract the most valuable information from the text. Second, the front-end deployment, constructed using Django, allows for end-users to easily access the summarized information without needing any programming ability.

Future Directions: SAMARA, like many web scrapers, suffers from difficulties when scraping non-standard pages and may incorrectly flag them as empty or not relevant. Though iGEM’s pages maintain relative standardization, it remains an issue when scraping hundreds of independently-created pages. This is being somewhat mitigated in other industries through the development of AI-based web scrapers, but the technology remains immature. SAMARA’s summarizer, the sshleifer distilbart cnn 12-6 model, is also imperfect. Though it produces acceptable results, the quality is not yet comparable to those of modern, GPT-3 based models. The main drawback of the GPT-3 models was the cost associated with their use, one that we avoided to remain as open source as possible. However, in the future, should these models be made open source and free to use, updating the summarizer would be a substantial upgrade in the quality of the summarized text.

Co-culture Modelling

Goal: The aim of the modeling project was to have an understanding of and validate the creation of a co-culture between K. xylinus and E. coli, inherently giving us explorations into the conditions that can allow for the optimal production of BC, PHB and Nisin.

Result: After creating mathematical models that consisted of a system of differential equations, we were able to run simulations that derived plots of the biomass concentration, substrate consumption, production of BC , Nisin and other extracellular substances that K. xylinus secretes which are acetic and gluconic acid. Ultimately, this helped inform wet lab experiments by giving us an understanding of how to regulate growth and environmental factors that influence production of our desired substances.

Future Directions: The implemented model falls short in some areas which can be resolved by further steps. One of such steps would involve conducting experiments to get data and use it to fit the model and improve it. Another improvement would involve considering other factors in the model that influence the behavior of the co-culture such as oxygen levels, surface area, and addition of other nutrients apart from the substrate consumed. Lastly, the model can be modified to implement more optimization techniques like Flux balance analysis to understand how we can make the maximum yield of our co-culture.

Uniaxial Testing

Goal: Uniaxial tensile testing method was used to explore BC as a viable alternative to existing packaging materials and provide valuable feedback to the wetlab design. Through uniaxial testing, mechanical properties of BC were quantitatively analyzed as the characteristic stress-strain curves of BC were generated to assess its competitive strength.

Result: As the purpose of the uniaxial tensile testing was to explore BC as a viable alternative as a packaging material, we can conclude from the collected data and analyzed mechanical properties that BC’s mechanical properties are customizable depending on the conditions of its culture and how it is treated.

Future Directions: Another important consideration is to validate our BC material using different data analysis. By utilizing different data analysis, we can investigate how the material will behave under different conditions, predict its failures, and improve the mechanical properties. Instead of solely testing its behavior by subjecting it to constant increase in strain until failure, we could also explore stress relaxation and cyclic loading to simulate different loading conditions.

Docking and Molecular Dynamics

Goal: The aim of conducting docking simulations was to understand the interactions and binding dynamics Nisin would have to both PHB and BC. Molecular dynamics simulations enabled us to understand the stability and conformational changes Nisin would undergo when put under several environmental conditions from production to use.

Result: Through docking simulations conducted on Nisin with BC, and Nisin with PHB, we found that Nisin has good affinity for these polymers. However, its affinity to BC is higher, thus validating that Nisin would immobilize well unto BC. We also discovered that the binding area of BC to Nisin is not the same as amino acid residues in Nisin that are effective against bacteria. Empirically, Nisin would be able to still perform its function after being immobilized unto BC.

Future Directions: Moving forward, molecular dynamics simulations on Nisin-BC and Nisin-PHB complexes that stem from docking simulations would be conducted so as to determine stability of the peptide when in this conformation. Also, conducting molecular dynamics will help us understand how these complexes change when several environmental conditions are in place such as changing temperatures, pressure or pH.