Introduction

We continuously engaged with experts and stakeholders to enhance our understanding of our work and gain valuable feedback on our project and its components.

Shaping the Problem

It was important that in validating our approach to advancing sustainable energy that we were careful and engaged with relevant stakeholders to validate that our approach does indeed have significant utility in advancing sustainable energy objectives. To do this we contacted:

• Dr. Daniel Bond, Professor of Microbiology at the University of Minnesota
• Dr. Bruce Logan, Professor of Civil and Environmental Engineering at Pennsylvania State University
• Dr. Douglas Call, Professor of Civil, Construction, and Environmental Engineering at North Carolina State University
• Dr. Sovik Das, Assistant Professor of Civil Engineering at the Indian Institute of Technology Delhi

The first iteration of our project focused on optimizing Microbial Fuel Cells (MFCs) as a method to use biological systems to generate electricity. The Cysteinator was originally intended to maintain the anoxic conditions in these cells. When asked to provide feedback on this original plan, Dr. Bond pointed out that the presence of a biological device would do more to maintain the anoxic conditions with its oxygen consumption than the L-cysteine that the device secreted. For this reason, he suggested that we pivot to looking into Microbial Electrolysis Cells (MECs) as an avenue to apply for our device. In response to his feedback, we researched how our device could be applied to MECs and made sure we evaluated the effect of the presence that our device has on the systems they are implemented into.

When seeking feedback from Professor Logan and Professor Call regarding our new MEC approach, we learned that MECs preferred acetate and alone were inadequate for converting waste substrates into hydrogen at a viable scale for large-scale implementation. Furthermore, the professors noted that pilot-scale MECs almost always produced methane rather than expected hydrogen. In response to their feedback, we sought information regarding the current state of MEC technologies and if progress had been made to resolve issues with hydrogen yields.

We reached out to Professor Das, who assured us that since discovering MECs, engineers have made numerous technological advancements to overcome these issues. We learned that through careful pretreatment methods, methanogenesis can be minimized. Most importantly, he suggested that we circumnavigate the MECs preference for acetate through a coupled system pairing MEC hydrogen production with dark fermentation. In response to his feedback, we realized that the Cysteinator’s capability to overproduce L-cysteine could be directly incorporated into these coupled systems, significantly optimizing dark fermentation feedstocks, and consequently increasing overall hydrogen production. This led us to identify our current problem, optimizing DF feedstocks.

Building the Cysteinator

When building the Cysteinator, we engaged with relevant researchers to validate our circuit design and help with our assembly. To do this, we contacted:

• Dr. Wei Liang, Associate Researcher at the Key Laboratory of Engineering Biology for Low-Carbon Manufacturing at the Institute of Industrial Biotechnology, Chinese Academy of Sciences, Tianjin, China
• Dr. Danielle Cook, Assistant Professor of Medicine, Division of Hematology & Oncology

When designing our device’s kill switch, we encountered problems with the complexity of modifying relevant regulatory elements. Our novel composite part took inspiration from Gao et al., who first took advantage of the L-cysteine responsive transcriptional regulator CcdR native to P. Ananatis (citation). We hoped to use the best-performing promoter and RBS that they identified, but we realized that given the overlap of the promoter and RBS of CcdR with the transcription factor binding region of a promoter of interest (pCcdA) in the same location on opposite strands of DNA, attempts to modify one element could lead to a reduction of function in the other. We decided to reach out to Dr. Liang, the lead author of the paper, for advice on navigating this issue. He agreed with our assumption that directly replacing the CcdR regulatory sites would be an unrealistic approach. He then recommended that we express CcdR with a promoter-RBS combination elsewhere on the plasmid, which would work so long as the CcdR binding site upstream of the CcdA promoter was preserved. In response to his feedback, we updated the design of our kill switch, awarding us full flexibility in tuning CcdR gene expression without interfering with the CcdR transcription factor binding site and native regulatory elements.

Building our device came with an assortment of challenges, specifically in our plasmid assembly strategy. We reached out to Dr. Cook, who has experience with in-fusion cloning, to help us troubleshoot our assembly. She gave us a multitude of suggestions on how to modify incubation times, PCR primers, and transformation protocols. Over the course of our assembly process, we maintained regular contact with Dr. Cook and continuously worked with her to troubleshoot our assembly processes. Building on her feedback, we were able to make significant progress on both understanding in-fusion assembly and assembling our plasmids. Further, Dr. Cook graciously assisted our team by donating CcdB survival cells and In-Fusion Snap Assembly Master Mix, which were critical to the success of our assembly.

Constructing the Model

When designing our multi-scale model, we consulted with experts to validate our novel design and inform our build. To do this we contacted:

• Dr. Pierre Salvy, Author of the Expression and Thermodynamic Flux (ETFL) model
• Dr. Zhitao Mao, Associate Researcher at the Institute of Industrial Biotechnology, Chinese Academy of Sciences, Tianjin
• Dr. Jason Papin, Professor of Biomedical Engineering at the University of Virginia
• Mr. Connor Moore, Graduate Student in the Papin Lab at the University of Virginia
• Dr. Matthew Peterson, Principal Data Scientist at the MITRE Corporation
• Dr. Jeff Saucerman, Professor of Biomedical Engineering at the University of Virginia

For constraint-based modeling, we initially chose to work with a genome scale metabolic model (GEM), specifically iML1515, and use the enzyme and thermodynamics flux (ETFL) framework to incorporate additional expression and thermodynamics data. The addition of omics data was intended to impose additional constraints while performing flux balance analysis (FBA) to increase the accuracy of the predicted optimal solution. We planned to implement this framework with the iML1515 model. However, we faced several challenges while trying to recreate the results from the ETFL paper using the code from ETFL’s repository. When facing difficulty in modifying the model, we reached out to Dr. Salvy, the author of the paper, who informed us that the ETFL code likely aged out of its practical functionality and is currently being refactored. Based on his feedback, we decided to pivot using the baseline GEM model on its own or attempt to incorporate a different framework/different type of omics data.

After further research, we decided it may be more accurate to instead incorporate enzyme constraints based on proteomics data to replace expression and thermodynamics data. We felt incorporating enzyme data would also be more relevant to our device design since many of the genetic modifications made are with mutating enzymes to increase their efficiency. We specifically chose to implement the ECMpy workflow to incorporate enzyme constraints. We reached out to Dr. Mao, the author of the ECMpy workflow, to determine the best way to reflect the metabolic changes our L-cysteine overproduction circuitry makes on the system. Dr. Mao detailed how we can edit the model's relevant CSV and JSON files, containing parameters related to building the enzyme constraint, to reflect modified enzyme constraints from the overexpression of the L-cysteine production mutants. Regarding characterizing L-cysteine transport, Dr. Mao informed us that the model was limited in its ability to characterize transport proteins. His feedback informed both the final construction of our FBA and the assumptions/limitations we had to keep in mind while forming conclusions based on the results from the model.

We faced difficulties while accurately reflecting modifications made in the genetic circuit in the model. To address these difficulties, we reached out to Dr. Papin, Dr. Peterson, and Mr. Moore, who provided insights regarding the technical aspects of building the model and setting parameters.

Dr. Papin gave us valuable insight into formulating the bounds for uptake reactions and determining which media components can be considered unlimited or limited. This was a large challenge we faced since concentrations of media components cannot be directly translated into uptake rates that are required for FBA. The FBA optimization function only allows for one reaction/pathway to be optimized. This posed another challenge since we aimed to optimize L-cysteine export while maintaining biomass growth. In response to this, he recommended that we use the OptKnock algorithm (citation) to be able to simultaneously optimize biomass growth and L-cysteine export. We responded to Dr. Papin’s feedback by using his metrics to set the upper bound on media uptake rates on the media conditions; however, a further review of the OptKnock algorithm revealed that its intended use-case did not match what we wanted it to do.

Dr. Peterson conversely suggested that we conduct a lexicographic optimization in which we first optimize for biomass growth, constrain that max growth using a conservative approximation and then optimize for L-cysteine export. We responded to Dr. Peterson’s feedback by implementing the lexicographic optimization.

When troubleshooting unexpected results from the fluxes predicted by FBA, Mr. Moore suggested performing a flux-variance analysis (FVA) to ensure that the expected pathways involved in L-cysteine production are carrying flux. flux through our engineered pathways is occurring. We responded to Mr. Moore’s feedback by conducting a flux-variance analysis to validate the fluxes were moving through the right pathways. Through this analysis, we found key reactions missing from the iML1515 model and were able to gapfill by incorporating the reaction and associated metabolites, essentially increasing the accuracy of the model for future use.

After designing the equations for the mechanistic model of the L-cysteine activated kill-switch we reached out to Dr. Saucerman to validate if our approach held merit and to ensure the assumptions we made while formulating the equations were reasonable to make. Dr. Saucerman confirmed our general approach was correct and offered suggestions on how to use modified Hill equations and integrate stochasticity into the model. We responded to his feedback by integrating his modified Hill equations. By integrating his feedback, we were able to also reduce the number of kinetic parameters needed. The hill-like approximations mainly required Kd values, which we were able to determine from molecular docking simulations. Reducing the number of approximations that would come with numerous kinetic parameters helped increase the accuracy of our model. We took inspiration from his view of integrating stochasticity into the model but ultimately chose to expand on his original idea for a more accurate and advanced model.

Read more about how we integrated this feedback on our modeling page.

Analyzing the UN Sustainable Development Goals

When conducting the meta-analysis of how the Cysteinator aligns with the UN Sustainable Development Goals (SDGs) we consulted with a wide variety of experts to more accurately gauge the true impact of our device. To analyze the 3 most relevant SDGs, we contacted:

• Dr. Cresten Mansfeldt, Professor of Environmental Engineering at the University of Colorado Boulder
• Dr. Michelle Halsted, Lead Biochemical Systems Engineer at Noblis
• Mr. Shakeel Balroop, Principal in Sustainable Waste Management and Circular Economy
• Dr. Sovik Das, Assistant Professor of Civil Engineering at the Indian Institute of Technology Delhi
• Dr. Lisa Colosi-Peterson, Professor of Civil and Environmental Engineering at the University of Virginia
• Dr. Andrés Clarens, Professor of Civil and Environmental Engineering at the University of Virginia
• Dr. Michael Lenox, Professor in Business Administration and Public Policy at the Batten School of Leadership and Public Policy
• Dr. William Shobe, Director Emeritus of the Center for Economic & Policy Studies
• Dr. Scott Doney, Professor and Marine Scientist at the University of Virginia, Department of Environmental Sciences
• Dr. Patrick O’Rourke, Post Doctorate Research Assistant at the University of Maryland College Park

To analyze SDG 6, which details available and sustainable water and sanitation, we spoke to Dr. Mansfeldt, Dr. Halsted, Mr. Balroop, Dr. Das, and Dr. Colosi-Peterson. They gave us valuable insight into how the Cysteinator can be integrated into waste-to-energy (WTE) streams and the impact that it can have on water quality, water-use efficiency, and wastewater treatment technologies. Read more about how we integrated their feedback on our SDG 6 page.

To analyze SDG 7, which details affordable, reliable, and sustainable energy, we spoke to Dr. Clarens and Dr. Lenox. They gave us valuable insight into how the Cysteinator can make sustainable energy more accessible and affordable. Read more about how we integrated their feedback on our SDG 7 page.

To analyze SDG 13, which details sustainable development, we spoke to Dr. Shobe, Dr. Colosi-Peterson, Dr. Doney, & Dr. O’Rourke. They gave us valuable insight into understanding what sustainable development is, its importance, and how to evaluate the Cysteinator’s potential to contributing to it. Read more about how we integrated their feedback on our SDG 13 page.

Considering Future Implications

Part 1: Hazard Prevention in the Modified DF Process Flow

When determining how our device will be implemented, we found it important to consider the biosafety of our system, both for us and the environment. To more expertly plan this, we contacted:

• Dr. Cresten Mansfeldt, Professor of Environmental Engineering at the University of Colorado Boulder.
• Dr. Benjamin Erable, Centre National de la Recherche Scientifique (CNRS) Research Manager at the Laboratoire de Genie Chimique, Toulouse, France

When considering the implementation of our device, we initially proposed the Cysteinator’s addition to bioreactors to occur simultaneously with the addition of waste streams utilized as feedstock. When we spoke with Professor Erable and Professor Mansfeldt, they both told us to assess our device’s growth capacity in bioreactors. Specifically, whether our bacteria could realistically coexist with the native microbial population. We realized we needed to evaluate whether other bacteria present would compete for resources and act as an antagonist to the Cysteinator, as well as understand threats our device might in turn pose to other bacteria—an example being the risk of horizontal gene transfer. In response to their feedback, we considered an alternative method of applying our device: instead of implementing the Cysteinator into feedstocks before pretreatment, we determined that adding the Cysteinator after pretreatment would minimize our devices interaction with other microbes. In addition to this we integrated an analysis of the Cysteinator’s impact on microbial communities into the future steps of our project. This will give us a more detailed understanding of the biosafety risks with our modified DF process flow.

Professor Mansfeldt noted that it is also important to consider how our modified process flow would mitigate the risk of our device propagating beyond the bioreactor and persisting past the DF process. He suggested to supplement our current biocontainment strategy, the L-cysteine activated kill-switch, with physical or chemical redundancy. In response to his feedback, we integrated an additional heating step into our process flow. Furthermore, we also investigated other genetic biocontainment tools.

Part 2: Applications to Other Industries

When determining future applications of the Cysteinator, we considered its potential to increase the sustainability of industrial processes as it related to biohydrogen production. Through our engagement with stakeholders, we discovered the broader applications of our device to a variety of industries. To gain more insight into these applications, we contacted:

• Dr. Andres Clarens, Professor of Civil and Environmental Engineering at the University of Virginia, Associate Director of the UVA Environmental Institute
• Dr. Ashty Karim, Assistant Professor of Chemical and Biological Engineering at Northwestern University

During our discussion with Professor Clarens, we learned that the Cysteinator could be impactful in the chemical sector. Specifically, he emphasized the Cysteinator has immense potential to mitigate greenhouse gas emission if adapted for use in the petrochemical sector’s bioreactors. Companies such as LanzaTech convert industrial carbon emissions to valuable products using anaerobic bacteria, which the Cysteinator can be appropriated to support. Our discussion with Professor Clarens prompted us to consider how other applications of the Cysteinator can contribute to sustainable development.

To gain further insight, we spoke to Professor Karim, who noted the immense opportunity for synthetic biology to contribute to addressing the climate crisis and the advancement of biohybrid systems: where biology meets other disciplines like chemistry and chemical engineering. The professor’s collective feedback enlightened us to future possibilities for the sustainable implementation of our device within the biomanufacturing and petrochemical industries.

Collaborations

SDG Instagram Post – VIT Vellore

We wanted to emphasize the importance of the UN Sustainable Development Goals (SDGs) and highlight how our project connects to them. To raise awareness about how synthetic biology can be used to advance SDGs, we collaborated with the VIT Vellore iGEM team on their initiative to create a series of Instagram posts showcasing each SDG and how iGEM teams are contributing solutions to their related challenges. Our team chose to focus on SDG 7, which aims to “ensure access to affordable, reliable, sustainable, and modern energy for all.” The Cysteinator helps to decrease the cost of H2 energy making it more sustainable and accessible, and this collaboration allowed us to bring global awareness to how we tackled SDG 7. We are confident that this increased awareness will spark other creative approaches to using synthetic biology in addressing SDG 7.

iGEM BIO ME Book – McGill

The BIOME book series, created by the McGill iGEM team, is an artistic initiative designed to communicate synthetic biology through art and visual storytelling. Each edition highlights the most important biological structures that iGEM teams incorporate into their projects. For this year's rendition of the BIOME book, McGill asked us to detail a molecule or protein central to our device’s gene circuit and signaling pathways. We chose to highlight the CcdB toxin, and discussed how it related to our favorite, infamous television chef personality, Gordon Ramsey! This creative exercise is aimed to be a fun way to encourage teams to consider how others use molecules and proteins as a crux to their project and help inform future synthetic biology decisions.

Learn more about the BIO ME Book on the McGill iGEM Wiki.

Mid-Atlantic Meetup

All iGEM teams in the Mid-Atlantic region of the U.S. were invited to the College of William and Mary on July 25th for a mini symposium organized by the William and Mary iGEM team. The University of Maryland, University of Virginia, and William and Mary iGEM teams presented their projects and shared their progress, creating an open space for collaboration and discussion. We had the opportunity to ask questions, learn about each other’s work, and receive valuable feedback from fellow teams and principal investigators and faculty from other teams. Professor Saha from the College of William and Mary gave us valuable feedback on constructing our wiki and being confident when we pivot, specifically in regard to our plasmid assembly strategies. We also had a great interaction with members from William and Mary team about finding and estimating parameters for modeling.

North American Mini Jamboree

At the North American Mini Jamboree on August 30th, our team joined other iGEM teams from across the continent in a virtual event hosted by groups including Boston University and the University of Calgary. We presented our project and received early feedback from judges, peers, and industry professionals, which helped us identify areas to refine before the Grand Jamboree. Their feedback was especially valuable in refining how we present the problem, and how the Cysteinator fits into the Dark Fermentation process flow. The event also allowed us to connect with other teams and synthetic biology professionals working on synthetic biology challenges, opening doors to collaboration opportunities and the sharing of ideas. In addition to presenting, our team attended workshops and speaker sessions that helped improve how we present, analyze, and build our project.