Introduction
In the process of developing and addressing our three project goals our team developed novel techniques, approaches, and learned valuable lessons. The page below details how other synthetic biologists can adapt our genetic circuitry, device testing, model development and testing, and approaches to analyzing device impact to enhance their own projects.
Integrated Modularity into Genetic Circuitry
Our intentional design of modularity into our genetic circuitry is an example of how intentional design can be responsive to ever-evolving diverse process flows. By being able to modify the circuitry in our device, the Cysteinator will be able to produce L-cysteine more quickly and at different concentrations. Allowing it to be used with multiple feedstocks and for different applications. Teams should consider integrating modularity into their circuitry so that synthetic biology solutions can be applied more broadly.
See our Engineering Page for a Detailed Explanation:
Novel Composite Part
The L-cysteine activated kill switch, is adapted from a bidirectional switch from P. ananatis. The CcdR protein that activates the kill switch acts both as repressor and a transcription factor. This part can be adapted for L-cysteine based regulation in a variety of applications. The use of this bidirectional switch enables more advanced & creative circuitry which will make other synthetic biology devices more efficacious.
Look at our Parts Page for a Detailed Explanation:
Ninhydrin Assay
Our ninhydrin assay protocol can be used to quantify amino acid concentration in a safe and effective manner compared to past iterations. Combined with the calibration curve our protocol details an effective way to convert qualitative color change into quantitative amino acid concentrations. This unique procedure will be able to advance the characterization of biological devices that utilize amino acids.
In addition to an updated protocol, our team encourages utilizing the following techniques when running ninhydrin assays or their calibration curves:
- Stagger the times at which samples begin boiling to reduce the time it takes to obtain OD560 readings.
- Have 2 people performing the ninhydrin assays. One person should take OD readings while the other prepares cuvettes. This decreases the time between when the cuvette is prepared and when the reading is taken, minimizing inaccuracies.
-
Ensure that the vector is completely linearized
- We suggest doing this by first conducting a longer digest with fresh enzymes and performing a gel extraction to increase purity.
- If adding homologous ends through PCR, ensure that PCR primers are generated to minimize risk of forming primer dimers, as they posed a significant hurdle to successful ligation.
-
Using Takara Stellar Competent Cells
- If forced to use another strain, dilute In-Fusion product 1:5 before transformation for optimal efficiency.
- Build in Backup Plans! The rate-limiting step on evaluating device efficacy data is often device assembly. Building in the ability to try multi assembly protocols creates redundancy and affords the opportunity to acquire Wet Lab data in a timely manner.
- Emphasize Record Keeping! Maintaining a lab notebook is critical to ensuring replicability and transparency. While all iGEM teams are required to maintain a lab notebook we learned that detailing weekly summaries of activities and their justifications is very beneficial for troubleshooting.
See our Experiments Page for a Detailed Explanation:
Modified Western Blot
We modified a traditional western blot protocol to conclusively determine if our L-cysteine activated kill-switch is functional. Specifically, we developed a calibration curve on each gel to assess the relationship between toxin concentration and fluorescence. This allowed us to determine with statistical significance whether our kill-switch was activated and contributing to precipitating cell death.
See our Experiments Page for a Detailed Explanation:
in-Fusion Assembly
Our plasmids were assembled using in-Fusion assembly [1]. in-Fusion assembly provides the benefits of a 15-minute reaction, no required subcloning step, high efficiency, seamless construction, and flexible experimental design. After troubleshooting the assembly process, we have developed suggestions for future teams conducting in-Fusion assembly:
See our Experiments Page for a Detailed Explanation:
Framework to Confer Stochasticity onto Mechanistic Models
To accurately represent the expression of CcdB we integrated a Markov Chain to build in stochasticity into the mechanistic model. Incorporating stochasticity was necessary because we cannot assume that CcdB will always be expressed due to relatively low transcription factor concentrations. This increased the accuracy of the mechanistic model by allowing us to run the model multiple times, simulating a population average. A single run of the model would not be able to confer the randomness of CcdB transcription action, and by using stochasticity on a multitude of runs we accurately predicted the behavior of the L-cysteine activated kill-switch.
This methodology should be adapted to other similar systems, and it can confer greater accuracy in mechanistic models. Future synthetic biologists should consider using stochasticity to account for randomness in cell systems and reduce the burden of traditional assumptions (eg. Quasi-Steady-State Reduction) on model accuracy.
See our Modeling Page for a Detailed Explanation:
Flux-Balance Analysis with Enzyme Constraints
The most accurate way to model devices that metabolically engineer strains is through a genome scale metabolic model (GEM). We ran a flux-balance analysis (FBA) using a GEM model to the maximum flux of metabolites through our engineered pathway. Furthermore, we acknowledged that running FBA on a baseline GEM has some limitations due to a large solution space. We addressed this issue by incorporating enzyme constraints to narrow the solution space and increase the accuracy of the predictions. The enzyme constraints essentially ensured that fluxes through pathways are capped by enzyme availability and the catalytic efficiency of the enzymes. Our methodology to reflect engineered changes in the constraint-based model and to run the FBA is a framework synthetic biologists can use to approximate maximum flux in metabolically engineered systems. The workflow can also be easily adapted for future use using the code linked in the team’s GitHub on the Modeling Page.
See our Modeling Page for a Detailed Explanation:
3D Structure Generation and Molecular Docking
Because the transcription factor that was used for the Kill-Switch was not well characterized in literature it became necessary to model its 3D structures for use in molecular docking simulations. Our method for developing and docking the 3D structure of the transcription factor, its dimer, tetramer, and octamer is an example of how parameters for novel components can be approximated. Furthermore, we used homology modeling to develop a 3D structure of the mutant CcdR monomer. Our workflow and 3D structures may be adapted for future use.
See our Modeling Page for a Detailed Explanation:
Iterative Model Optimization
Our project employs a strategic iterative model optimization framework wherein results from benchwork inform model and vice versa. This framework ensures that both the Cysteinator and our model of it improve together over time, improving their joint efficacy and leading to increased hydrogen yields. This framework is an example for computational and synthetic biologists on how to effectively integrate their work.
Look at our Engineering Page for a Detailed Explanation:
Figure 4. Figure 1. The process flow for how the genetic circuit design, model, wet lab experimentation, and industrial applications interact. Blue arrows depict the flow data originating from lab experimentation informing iterative design. Orange arrows depict the flow of data originating from the model informing iterative design.
Life-Cycle Analysis and Techno-Economic Analysis
We conducted a life-cycle analysis (LCA) and a techno-economic analysis (TEA) to ascertain the impact of the Cysteinator on DF-based hydrogen production. These analyses gave us vital information and context about the benefits and pitfalls of our device in regard to sustainable development, as well as helped constrain the future development of the Cysteinator. By adapting previous LCA and TEAs, we were able quantify comparisons between our device and the status quo. Conducting an LCA and TEA, significant tools of environmental and economic analysis, can thus help synthetic biologists better understand the need for innovation in sustainable energy production.
See our SDG Page for a Detailed Explanation:
Advice for Future Teams
After reflecting on our experience conducting our project, we suggest future iGEM teams:
Future Work
The immediate future work is to continue wet lab testing of the L-cysteine overproduction circuit, the L-cysteine activated kill-switch circuitry, and the combination of them both. Based on the results of this work and our model, iterative design will occur to optimize the Cysteinator and its model (see Contribution 9).
Once the Cysteinator secretes L-cysteine to a concentration of 300 mg/L in culture media, we will test its ability to do the same in synthetic wastewaters feedstocks. We will then test H2 production from these feedstocks in line with precedence [2]. Also tested will be the growth of hydrogen-producing bacteria and the reduction of the oxidative-reductive potential. Data from this stage of testing will continue to refine the model.
When proven functional in synthetic feedstocks, future work will be repeat the previous experiments in traditional feedstocks. In addition to factors being tested above, the biodiversity of feedstock originated inoculum will also be evaluated. This will be done with multiple feedstocks to determine the type of feedstock the Cysteinator is most successful in.
Once this is done, a full-scale test of DF bioreactor will have the Cysteinator incorporated into its process flow. All experiments from the previous testing round will be repeated, and the data used to perform final calibrations on the model.
All along this process, data from both the wetlab and model will be used to continuously evaluate how iterations the Cysteinator will impact the sustainable energy transition.